A Complete List of All (arXiv) Adversarial Example Papers
by Nicholas Carlini 2019-06-15
It can be hard to stay up-to-date on the published papers in the field of adversarial examples, where we have seen massive growth in the number of papers written each year. I have been somewhat religiously keeping track of these papers for the last few years, and realized it may be helpful for others to release this list.
The only requirement I used for selecting papers for this list is that it is primarily a paper about adversarial examples, or extensively uses adversarial examples. Due to the sheer quantity of papers, I can't guarantee that I actually have found all of them.
But I did try.
I also may have included papers that don't match these criteria (and are about something different instead), or made inconsistent judgement calls as to whether or not any given paper is mainly an adversarial example paper. Send me an email if something is wrong and I'll correct it.
As a result, this list is completely un-filtered. Everything that mainly presents itself as an adversarial example paper is listed here; I pass no judgement of quality. For a curated list of papers that I think are excellent and worth reading, see the Adversarial Machine Learning Reading List.
One final note about the data. This list automatically updates with new papers, even before I get a chance to manually filter through them. I do this filtering roughly twice a week, and it's then that I'll remove the ones that aren't related to adversarial examples. As a result, there may be some false positives on the most recent few entries. The new un-verified entries will have a probability indicated that my simplistic (but reasonably well calibrated) bag-of-words classifier believes the given paper is actually about adversarial examples.
The full paper list appears below. I've also released a TXT file (and a TXT file with abstracts) and a JSON file with the same data. If you do anything interesting with this data I'd be happy to hear from you what it was.
Paper List
2024-12-02
Traversing the Subspace of Adversarial Patches. (83%)
DiffPatch: Generating Customizable Adversarial Patches using Diffusion Model. (82%)
Exploring the Robustness of AI-Driven Tools in Digital Forensics: A Preliminary Study. (74%)
Adversarial Sample-Based Approach for Tighter Privacy Auditing in Final Model-Only Scenarios. (69%)
Robust and Transferable Backdoor Attacks Against Deep Image Compression With Selective Frequency Prior. (67%)
Adversarial Attacks on Hyperbolic Networks. (26%)
CopyrightShield: Spatial Similarity Guided Backdoor Defense against Copyright Infringement in Diffusion Models. (10%)
R.I.P.: A Simple Black-box Attack on Continual Test-time Adaptation. (5%)
Precision Profile Pollution Attack on Sequential Recommenders via Influence Function. (1%)
2024-12-01
Intermediate Outputs Are More Sensitive Than You Think. (61%)
Hiding Faces in Plain Sight: Defending DeepFakes by Disrupting Face Detection. (16%)
Online Poisoning Attack Against Reinforcement Learning under Black-box Environments. (11%)
2024-11-30
Hard-Label Black-Box Attacks on 3D Point Clouds. (99%)
Exposing LLM Vulnerabilities: Adversarial Scam Detection and Performance. (69%)
Exact Certification of (Graph) Neural Networks Against Label Poisoning. (22%)
Jailbreak Large Vision-Language Models Through Multi-Modal Linkage. (12%)
2024-11-29
Towards Class-wise Robustness Analysis. (99%)
FLARE: Towards Universal Dataset Purification against Backdoor Attacks. (81%)
Robust Table Integration in Data Lakes. (56%)
On the Adversarial Robustness of Instruction-Tuned Large Language Models for Code. (38%)
Parallel Stacked Aggregated Network for Voice Authentication in IoT-Enabled Smart Devices. (10%)
Fusing Physics-Driven Strategies and Cross-Modal Adversarial Learning: Toward Multi-Domain Applications. (1%)
SURE-VQA: Systematic Understanding of Robustness Evaluation in Medical VQA Tasks. (1%)
2024-11-28
SceneTAP: Scene-Coherent Typographic Adversarial Planner against Vision-Language Models in Real-World Environments. (84%)
PEFT-as-an-Attack! Jailbreaking Language Models during Federated Parameter-Efficient Fine-Tuning. (69%)
Random Sampling for Diffusion-based Adversarial Purification. (26%)
Understanding and Improving Training-Free AI-Generated Image Detections with Vision Foundation Models. (11%)
LADDER: Multi-objective Backdoor Attack via Evolutionary Algorithm. (2%)
Enhancing Neural Network Robustness Against Fault Injection Through Non-linear Weight Transformations. (2%)
2024-11-27
Visual Adversarial Attack on Vision-Language Models for Autonomous Driving. (99%)
Fall Leaf Adversarial Attack on Traffic Sign Classification. (99%)
Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment. (67%)
Neutralizing Backdoors through Information Conflicts for Large Language Models. (26%)
Hidden Data Privacy Breaches in Federated Learning. (22%)
SoK: Watermarking for AI-Generated Content. (3%)
From Open Vocabulary to Open World: Teaching Vision Language Models to Detect Novel Objects. (1%)
2024-11-26
Adversarial Training in Low-Label Regimes with Margin-Based Interpolation. (99%)
BadScan: An Architectural Backdoor Attack on Visual State Space Models. (98%)
Stealthy Multi-Task Adversarial Attacks. (92%)
Adversarial Bounding Boxes Generation (ABBG) Attack against Visual Object Trackers. (82%)
MADE: Graph Backdoor Defense with Masked Unlearning. (82%)
Exploring Visual Vulnerabilities via Multi-Loss Adversarial Search for Jailbreaking Vision-Language Models. (75%)
Privacy-preserving Robotic-based Multi-factor Authentication Scheme for Secure Automated Delivery System. (9%)
PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning. (2%)
Multi-Objective Reinforcement Learning for Automated Resilient Cyber Defence. (1%)
Improved Parallel Derandomization via Finite Automata with Applications. (1%)
2024-11-25
Unlocking The Potential of Adaptive Attacks on Diffusion-Based Purification. (99%)
Imperceptible Adversarial Examples in the Physical World. (99%)
Scaling Laws for Black box Adversarial Attacks. (99%)
Privacy Protection in Personalized Diffusion Models via Targeted Cross-Attention Adversarial Attack. (81%)
UVCG: Leveraging Temporal Consistency for Universal Video Protection. (54%)
Guarding the Gate: ConceptGuard Battles Concept-Level Backdoors in Concept Bottleneck Models. (50%)
Edit Away and My Face Will not Stay: Personal Biometric Defense against Malicious Generative Editing. (50%)
Sparse patches adversarial attacks via extrapolating point-wise information. (47%)
DeDe: Detecting Backdoor Samples for SSL Encoders via Decoders. (10%)
RED: Robust Environmental Design. (10%)
BadSFL: Backdoor Attack against Scaffold Federated Learning. (3%)
Why the Agent Made that Decision: Explaining Deep Reinforcement Learning with Vision Masks. (2%)
XAI and Android Malware Models. (2%)
Revisiting Marr in Face: The Building of 2D--2.5D--3D Representations in Deep Neural Networks. (1%)
2024-11-24
Chain of Attack: On the Robustness of Vision-Language Models Against Transfer-Based Adversarial Attacks. (99%)
ExAL: An Exploration Enhanced Adversarial Learning Algorithm. (92%)
A Tunable Despeckling Neural Network Stabilized via Diffusion Equation. (64%)
Hide in Plain Sight: Clean-Label Backdoor for Auditing Membership Inference. (10%)
DRIVE: Dual-Robustness via Information Variability and Entropic Consistency in Source-Free Unsupervised Domain Adaptation. (2%)
Stealth Attacks Against Moving Target Defense for Smart Grid. (2%)
2024-11-23
Improving Transferable Targeted Attacks with Feature Tuning Mixup. (99%)
Enhancing the Transferability of Adversarial Attacks on Face Recognition with Diverse Parameters Augmentation. (99%)
Semantic Shield: Defending Vision-Language Models Against Backdooring and Poisoning via Fine-grained Knowledge Alignment. (4%)
LoBAM: LoRA-Based Backdoor Attack on Model Merging. (2%)
2024-11-22
Exploring the Robustness and Transferability of Patch-Based Adversarial Attacks in Quantized Neural Networks. (99%)
Gradient Masking All-at-Once: Ensemble Everything Everywhere Is Not Robust. (99%)
Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbreaks. (98%)
Derivative-Free Diffusion Manifold-Constrained Gradient for Unified XAI. (45%)
Who Can Withstand Chat-Audio Attacks? An Evaluation Benchmark for Large Language Models. (41%)
Universal and Context-Independent Triggers for Precise Control of LLM Outputs. (31%)
Benchmarking the Robustness of Optical Flow Estimation to Corruptions. (13%)
Twin Trigger Generative Networks for Backdoor Attacks against Object Detection. (4%)
Geminio: Language-Guided Gradient Inversion Attacks in Federated Learning. (2%)
Heavy-tailed Contamination is Easier than Adversarial Contamination. (1%)
Exploiting Watermark-Based Defense Mechanisms in Text-to-Image Diffusion Models for Unauthorized Data Usage. (1%)
Reliable Evaluation of Attribution Maps in CNNs: A Perturbation-Based Approach. (1%)
2024-11-21
Generating Realistic Adversarial Examples for Business Processes using Variational Autoencoders. (99%)
Learning Fair Robustness via Domain Mixup. (81%)
GASP: Efficient Black-Box Generation of Adversarial Suffixes for Jailbreaking LLMs. (78%)
Adversarial Prompt Distillation for Vision-Language Models. (75%)
AnywhereDoor: Multi-Target Backdoor Attacks on Object Detection. (74%)
Indiscriminate Disruption of Conditional Inference on Multivariate Gaussians. (50%)
GraphTheft: Quantifying Privacy Risks in Graph Prompt Learning. (4%)
Global Challenge for Safe and Secure LLMs Track 1. (4%)
TrojanEdit: Backdooring Text-Based Image Editing Models. (3%)
Evaluating the Robustness of Analogical Reasoning in Large Language Models. (1%)
Memory Backdoor Attacks on Neural Networks. (1%)
2024-11-20
Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attacks. (98%)
TAPT: Test-Time Adversarial Prompt Tuning for Robust Inference in Vision-Language Models. (96%)
Provably Efficient Action-Manipulation Attack Against Continuous Reinforcement Learning. (86%)
A Survey on Adversarial Robustness of LiDAR-based Machine Learning Perception in Autonomous Vehicles. (86%)
Rethinking the Intermediate Features in Adversarial Attacks: Misleading Robotic Models via Adversarial Distillation. (68%)
AI-generated Image Detection: Passive or Watermark? (22%)
SoK: A Systems Perspective on Compound AI Threats and Countermeasures. (12%)
CopyrightMeter: Revisiting Copyright Protection in Text-to-image Models. (12%)
Bounding-box Watermarking: Defense against Model Extraction Attacks on Object Detectors. (5%)
2024-11-19
NMT-Obfuscator Attack: Ignore a sentence in translation with only one word. (99%)
Stochastic BIQA: Median Randomized Smoothing for Certified Blind Image Quality Assessment. (75%)
When Backdoors Speak: Understanding LLM Backdoor Attacks Through Model-Generated Explanations. (3%)
2024-11-18
Theoretical Corrections and the Leveraging of Reinforcement Learning to Enhance Triangle Attack. (99%)
Adapting to Cyber Threats: A Phishing Evolution Network (PEN) Framework for Phishing Generation and Analyzing Evolution Patterns using Large Language Models. (87%)
DeTrigger: A Gradient-Centric Approach to Backdoor Attack Mitigation in Federated Learning. (75%)
CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization. (67%)
Reliable Poisoned Sample Detection against Backdoor Attacks Enhanced by Sharpness Aware Minimization. (50%)
Few-shot Model Extraction Attacks against Sequential Recommender Systems. (38%)
CLUE-MARK: Watermarking Diffusion Models using CLWE. (26%)
The Dark Side of Trust: Authority Citation-Driven Jailbreak Attacks on Large Language Models. (13%)
Exploring adversarial robustness of JPEG AI: methodology, comparison and new methods. (8%)
2024-11-17
Exploring the Adversarial Vulnerabilities of Vision-Language-Action Models in Robotics. (86%)
JailbreakLens: Interpreting Jailbreak Mechanism in the Lens of Representation and Circuit. (47%)
Countering Backdoor Attacks in Image Recognition: A Survey and Evaluation of Mitigation Strategies. (22%)
SoK: Unifying Cybersecurity and Cybersafety of Multimodal Foundation Models with an Information Theory Approach. (9%)
CLMIA: Membership Inference Attacks via Unsupervised Contrastive Learning. (2%)
2024-11-15
A Hard-Label Cryptanalytic Extraction of Non-Fully Connected Deep Neural Networks using Side-Channel Attacks. (98%)
Edge-Only Universal Adversarial Attacks in Distributed Learning. (98%)
Prompt-Guided Environmentally Consistent Adversarial Patch. (82%)
Continual Adversarial Reinforcement Learning (CARL) of False Data Injection detection: forgetting and explainability. (81%)
EveGuard: Defeating Vibration-based Side-Channel Eavesdropping with Audio Adversarial Perturbations. (68%)
Comparing Robustness Against Adversarial Attacks in Code Generation: LLM-Generated vs. Human-Written. (68%)
Safe Text-to-Image Generation: Simply Sanitize the Prompt Embedding. (11%)
Measuring Non-Adversarial Reproduction of Training Data in Large Language Models. (9%)
Toward Robust and Accurate Adversarial Camouflage Generation against Vehicle Detectors. (1%)
RedTest: Towards Measuring Redundancy in Deep Neural Networks Effectively. (1%)
2024-11-14
BEARD: Benchmarking the Adversarial Robustness for Dataset Distillation. (99%)
Transferable Adversarial Attacks against ASR. (89%)
Adversarial Attacks Using Differentiable Rendering: A Survey. (83%)
Jailbreak Attacks and Defenses against Multimodal Generative Models: A Survey. (69%)
Your Fixed Watermark is Fragile: Towards Semantic-Aware Watermark for EaaS Copyright Protection. (11%)
Are nuclear masks all you need for improved out-of-domain generalisation? A closer look at cancer classification in histopathology. (1%)
2024-11-13
Confidence-aware Denoised Fine-tuning of Off-the-shelf Models for Certified Robustness. (95%)
Trap-MID: Trapdoor-based Defense against Model Inversion Attacks. (81%)
Robust Optimal Power Flow Against Adversarial Attacks: A Tri-Level Optimization Approach. (81%)
The VLLM Safety Paradox: Dual Ease in Jailbreak Attack and Defense. (22%)
LLMStinger: Jailbreaking LLMs using RL fine-tuned LLMs. (8%)
2024-11-12
Chain Association-based Attacking and Shielding Natural Language Processing Systems. (99%)
IAE: Irony-based Adversarial Examples for Sentiment Analysis Systems. (99%)
Zer0-Jack: A Memory-efficient Gradient-based Jailbreaking Method for Black-box Multi-modal Large Language Models. (78%)
Deceiving Question-Answering Models: A Hybrid Word-Level Adversarial Approach. (67%)
A Survey on Adversarial Machine Learning for Code Data: Realistic Threats, Countermeasures, and Interpretations. (64%)
New Emerged Security and Privacy of Pre-trained Model: a Survey and Outlook. (13%)
Adaptive Meta-Learning for Robust Deepfake Detection: A Multi-Agent Framework to Data Drift and Model Generalization. (1%)
2024-11-11
Boosting the Targeted Transferability of Adversarial Examples via Salient Region & Weighted Feature Drop. (99%)
Computable Model-Independent Bounds for Adversarial Quantum Machine Learning. (69%)
The Inherent Adversarial Robustness of Analog In-Memory Computing. (61%)
Rapid Response: Mitigating LLM Jailbreaks with a Few Examples. (54%)
Semi-Truths: A Large-Scale Dataset of AI-Augmented Images for Evaluating Robustness of AI-Generated Image detectors. (1%)
2024-11-10
Adversarial Detection with a Dynamically Stable System. (99%)
Deferred Backdoor Functionality Attacks on Deep Learning Models. (82%)
SequentialBreak: Large Language Models Can be Fooled by Embedding Jailbreak Prompts into Sequential Prompt Chains. (70%)
2024-11-09
Target-driven Attack for Large Language Models. (73%)
AI-Compass: A Comprehensive and Effective Multi-module Testing Tool for AI Systems. (33%)
2024-11-08
Post-Hoc Robustness Enhancement in Graph Neural Networks with Conditional Random Fields. (41%)
Reasoning Robustness of LLMs to Adversarial Typographical Errors. (13%)
Towards a Re-evaluation of Data Forging Attacks in Practice. (2%)
2024-11-07
Neural Fingerprints for Adversarial Attack Detection. (99%)
Adversarial Robustness of In-Context Learning in Transformers for Linear Regression. (98%)
Seeing is Deceiving: Exploitation of Visual Pathways in Multi-Modal Language Models. (97%)
Attention Masks Help Adversarial Attacks to Bypass Safety Detectors. (97%)
Defending Deep Regression Models against Backdoor Attacks. (78%)
Hardware and Software Platform Inference. (5%)
Saliency Assisted Quantization for Neural Networks. (1%)
MISGUIDE: Security-Aware Attack Analytics for Smart Grid Load Frequency Control. (1%)
2024-11-06
Game-Theoretic Defenses for Robust Conformal Prediction Against Adversarial Attacks in Medical Imaging. (95%)
Deferred Poisoning: Making the Model More Vulnerable via Hessian Singularization. (86%)
FedRISE: Rating Induced Sign Election of Gradients for Byzantine Tolerant Federated Aggregation. (41%)
MRJ-Agent: An Effective Jailbreak Agent for Multi-Round Dialogue. (10%)
Towards Secured Smart Grid 2.0: Exploring Security Threats, Protection Models, and Challenges. (4%)
Mitigating Privacy Risks in LLM Embeddings from Embedding Inversion. (1%)
2024-11-05
Region-Guided Attack on the Segment Anything Model (SAM). (99%)
Enhancing Adversarial Robustness via Uncertainty-Aware Distributional Adversarial Training. (99%)
Double Whammy: Stealthy Data Manipulation aided Reconstruction Attack on Graph Federated Learning. (91%)
Flashy Backdoor: Real-world Environment Backdoor Attack on SNNs with DVS Cameras. (75%)
Formal Logic-guided Robust Federated Learning against Poisoning Attacks. (68%)
Oblivious Defense in ML Models: Backdoor Removal without Detection. (15%)
DM4Steal: Diffusion Model For Link Stealing Attack On Graph Neural Networks. (13%)
FEDLAD: Federated Evaluation of Deep Leakage Attacks and Defenses. (9%)
Lost in Context: The Influence of Context on Feature Attribution Methods for Object Recognition. (3%)
Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset. (1%)
2024-11-04
Query-Efficient Adversarial Attack Against Vertical Federated Graph Learning. (99%)
Semantic-Aligned Adversarial Evolution Triangle for High-Transferability Vision-Language Attack. (99%)
LiDAttack: Robust Black-box Attack on LiDAR-based Object Detection. (99%)
Alignment-Based Adversarial Training (ABAT) for Improving the Robustness and Accuracy of EEG-Based BCIs. (91%)
Attacking Vision-Language Computer Agents via Pop-ups. (9%)
FactTest: Factuality Testing in Large Language Models with Statistical Guarantees. (1%)
Stochastic Monkeys at Play: Random Augmentations Cheaply Break LLM Safety Alignment. (1%)
Differentially Private Integrated Decision Gradients (IDG-DP) for Radar-based Human Activity Recognition. (1%)
2024-11-03
Undermining Image and Text Classification Algorithms Using Adversarial Attacks. (98%)
SQL Injection Jailbreak: a structural disaster of large language models. (78%)
Rotation Perturbation Robustness in Point Cloud Analysis: A Perspective of Manifold Distillation. (2%)
TabSec: A Collaborative Framework for Novel Insider Threat Detection. (1%)
Learning predictable and robust neural representations by straightening image sequences. (1%)
2024-11-02
$B^4$: A Black-Box Scrubbing Attack on LLM Watermarks. (75%)
What Features in Prompts Jailbreak LLMs? Investigating the Mechanisms Behind Attacks. (1%)
2024-11-01
Replace-then-Perturb: Targeted Adversarial Attacks With Visual Reasoning for Vision-Language Models. (99%)
Defense Against Prompt Injection Attack by Leveraging Attack Techniques. (81%)
Certified Robustness for Deep Equilibrium Models via Serialized Random Smoothing. (68%)
Attention Tracker: Detecting Prompt Injection Attacks in LLMs. (26%)
Emoji Attack: A Method for Misleading Judge LLMs in Safety Risk Detection. (22%)
Outlier-Oriented Poisoning Attack: A Grey-box Approach to Disturb Decision Boundaries by Perturbing Outliers in Multiclass Learning. (13%)
Identify Backdoored Model in Federated Learning via Individual Unlearning. (5%)
Plentiful Jailbreaks with String Compositions. (2%)
Uncertainty-based Offline Variational Bayesian Reinforcement Learning for Robustness under Diverse Data Corruptions. (2%)
Examining Attacks on Consensus and Incentive Systems in Proof-of-Work Blockchains: A Systematic Literature Review. (1%)
B-cosification: Transforming Deep Neural Networks to be Inherently Interpretable. (1%)
Towards Building Secure UAV Navigation with FHE-aware Knowledge Distillation. (1%)
2024-10-31
Noise as a Double-Edged Sword: Reinforcement Learning Exploits Randomized Defenses in Neural Networks. (99%)
Protecting Feed-Forward Networks from Adversarial Attacks Using Predictive Coding. (99%)
Wide Two-Layer Networks can Learn from Adversarial Perturbations. (98%)
DiffPAD: Denoising Diffusion-based Adversarial Patch Decontamination. (93%)
I Can Hear You: Selective Robust Training for Deepfake Audio Detection. (86%)
Pseudo-Conversation Injection for LLM Goal Hijacking. (75%)
ARQ: A Mixed-Precision Quantization Framework for Accurate and Certifiably Robust DNNs. (41%)
Optical Lens Attack on Monocular Depth Estimation for Autonomous Driving. (5%)
Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey. (2%)
2024-10-30
FAIR-TAT: Improving Model Fairness Using Targeted Adversarial Training. (99%)
Keep on Swimming: Real Attackers Only Need Partial Knowledge of a Multi-Model System. (99%)
CausalDiff: Causality-Inspired Disentanglement via Diffusion Model for Adversarial Defense. (99%)
One Prompt to Verify Your Models: Black-Box Text-to-Image Models Verification via Non-Transferable Adversarial Attacks. (98%)
Effective and Efficient Adversarial Detection for Vision-Language Models via A Single Vector. (87%)
HijackRAG: Hijacking Attacks against Retrieval-Augmented Large Language Models. (82%)
Understanding and Improving Adversarial Collaborative Filtering for Robust Recommendation. (67%)
Teaching a Language Model to Distinguish Between Similar Details using a Small Adversarial Training Set. (64%)
Backdoor Attack Against Vision Transformers via Attention Gradient-Based Image Erosion. (62%)
Geometry Cloak: Preventing TGS-based 3D Reconstruction from Copyrighted Images. (2%)
Byzantine-Robust Federated Learning: An Overview With Focus on Developing Sybil-based Attacks to Backdoor Augmented Secure Aggregation Protocols. (1%)
ProTransformer: Robustify Transformers via Plug-and-Play Paradigm. (1%)
Attribute-to-Delete: Machine Unlearning via Datamodel Matching. (1%)
Stealing User Prompts from Mixture of Experts. (1%)
InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models. (1%)
2024-10-29
On the Robustness of Adversarial Training Against Uncertainty Attacks. (99%)
CausAdv: A Causal-based Framework for Detecting Adversarial Examples. (99%)
Text-Guided Attention is All You Need for Zero-Shot Robustness in Vision-Language Models. (98%)
IDEATOR: Jailbreaking Large Vision-Language Models Using Themselves. (83%)
Automated Trustworthiness Oracle Generation for Machine Learning Text Classifiers. (82%)
Longitudinal Mammogram Exam-based Breast Cancer Diagnosis Models: Vulnerability to Adversarial Attacks. (81%)
AmpleGCG-Plus: A Strong Generative Model of Adversarial Suffixes to Jailbreak LLMs with Higher Success Rates in Fewer Attempts. (78%)
Embedding-based classifiers can detect prompt injection attacks. (64%)
Enhancing Adversarial Attacks through Chain of Thought. (54%)
Power side-channel leakage localization through adversarial training of deep neural networks. (11%)
Enhancing Safety and Robustness of Vision-Based Controllers via Reachability Analysis. (1%)
DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Models. (1%)
2024-10-28
Evaluating the Robustness of LiDAR Point Cloud Tracking Against Adversarial Attack. (99%)
AdvI2I: Adversarial Image Attack on Image-to-Image Diffusion models. (96%)
BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks. (93%)
FATH: Authentication-based Test-time Defense against Indirect Prompt Injection Attacks. (91%)
TACO: Adversarial Camouflage Optimization on Trucks to Fool Object Detectors. (88%)
Attacking Misinformation Detection Using Adversarial Examples Generated by Language Models. (83%)
Hacking Back the AI-Hacker: Prompt Injection as a Defense Against LLM-driven Cyberattacks. (80%)
Stealthy Jailbreak Attacks on Large Language Models via Benign Data Mirroring. (50%)
Mitigating Unauthorized Speech Synthesis for Voice Protection. (9%)
Palisade -- Prompt Injection Detection Framework. (1%)
2024-10-27
Integrating uncertainty quantification into randomized smoothing based robustness guarantees. (98%)
LLM Robustness Against Misinformation in Biomedical Question Answering. (80%)
Fine-tuned Large Language Models (LLMs): Improved Prompt Injection Attacks Detection. (1%)
2024-10-26
Adversarial Attacks Against Double RIS-Assisted MIMO Systems-based Autoencoder in Finite-Scattering Environments. (99%)
Transferable Adversarial Attacks on SAM and Its Downstream Models. (99%)
Generative Adversarial Patches for Physical Attacks on Cross-Modal Pedestrian Re-Identification. (98%)
CodePurify: Defend Backdoor Attacks on Neural Code Models via Entropy-based Purification. (76%)
Robust Model Evaluation over Large-scale Federated Networks. (2%)
2024-10-25
GPT-4o System Card. (76%)
RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction. (64%)
Attacks against Abstractive Text Summarization Models through Lead Bias and Influence Functions. (62%)
Expose Before You Defend: Unifying and Enhancing Backdoor Defenses via Exposed Models. (56%)
Towards Robust Algorithms for Surgical Phase Recognition via Digital Twin-based Scene Representation. (2%)
2024-10-24
GADT: Enhancing Transferable Adversarial Attacks through Gradient-guided Adversarial Data Transformation. (99%)
Adversarial Attacks on Large Language Models Using Regularized Relaxation. (98%)
Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities. (88%)
Humanizing the Machine: Proxy Attacks to Mislead LLM Detectors. (68%)
Complexity Matters: Effective Dimensionality as a Measure for Adversarial Robustness. (33%)
Robust Watermarking Using Generative Priors Against Image Editing: From Benchmarking to Advances. (11%)
2024-10-23
Advancing NLP Security by Leveraging LLMs as Adversarial Engines. (98%)
Backdoor in Seconds: Unlocking Vulnerabilities in Large Pre-trained Models via Model Editing. (93%)
Breaking the Illusion: Real-world Challenges for Adversarial Patches in Object Detection. (70%)
Slot: Provenance-Driven APT Detection through Graph Reinforcement Learning. (16%)
Guide for Defense (G4D): Dynamic Guidance for Robust and Balanced Defense in Large Language Models. (9%)
Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks. (2%)
Countering Autonomous Cyber Threats. (2%)
Is Smoothness the Key to Robustness? A Comparison of Attention and Convolution Models Using a Novel Metric. (1%)
2024-10-22
Detecting Adversarial Examples. (99%)
Test-time Adversarial Defense with Opposite Adversarial Path and High Attack Time Cost. (98%)
AdvWeb: Controllable Black-box Attacks on VLM-powered Web Agents. (97%)
Meta Stackelberg Game: Robust Federated Learning against Adaptive and Mixed Poisoning Attacks. (67%)
Hierarchical Multi-agent Reinforcement Learning for Cyber Network Defense. (41%)
On the Vulnerability of Text Sanitization. (8%)
Context-aware Prompt Tuning: Advancing In-Context Learning with Adversarial Methods. (5%)
Evaluating the Effectiveness of Attack-Agnostic Features for Morphing Attack Detection. (4%)
BadFair: Backdoored Fairness Attacks with Group-conditioned Triggers. (2%)
Invisible Manipulation Deep Reinforcement Learning Enhanced Stealthy Attacks on Battery Energy Management Systems. (1%)
A Hybrid Simulation of DNN-based Gray Box Models. (1%)
2024-10-21
Model Mimic Attack: Knowledge Distillation for Provably Transferable Adversarial Examples. (99%)
Conflict-Aware Adversarial Training. (70%)
Robust Feature Learning for Multi-Index Models in High Dimensions. (68%)
Dual-Model Defense: Safeguarding Diffusion Models from Membership Inference Attacks through Disjoint Data Splitting. (16%)
Metric as Transform: Exploring beyond Affine Transform for Interpretable Neural Network. (13%)
A Realistic Threat Model for Large Language Model Jailbreaks. (11%)
Vulnerabilities in Machine Learning-Based Voice Disorder Detection Systems. (11%)
On the Geometry of Regularization in Adversarial Training: High-Dimensional Asymptotics and Generalization Bounds. (5%)
Boosting Jailbreak Transferability for Large Language Models. (1%)
Extracting Spatiotemporal Data from Gradients with Large Language Models. (1%)
2024-10-20
PEAS: A Strategy for Crafting Transferable Adversarial Examples. (99%)
Efficient Model Extraction via Boundary Sampling. (96%)
The Best Defense is a Good Offense: Countering LLM-Powered Cyberattacks. (76%)
Faster-GCG: Efficient Discrete Optimization Jailbreak Attacks against Aligned Large Language Models. (45%)
Bayesian Concept Bottleneck Models with LLM Priors. (1%)
2024-10-19
Adversarial Training: A Survey. (97%)
Toward Robust RALMs: Revealing the Impact of Imperfect Retrieval on Retrieval-Augmented Language Models. (92%)
Beyond Pruning Criteria: The Dominant Role of Fine-Tuning and Adaptive Ratios in Neural Network Robustness. (76%)
Jailbreaking and Mitigation of Vulnerabilities in Large Language Models. (50%)
SLIC: Secure Learned Image Codec through Compressed Domain Watermarking to Defend Image Manipulation. (11%)
DynaMO: Protecting Mobile DL Models through Coupling Obfuscated DL Operators. (2%)
2024-10-18
A Hybrid Defense Strategy for Boosting Adversarial Robustness in Vision-Language Models. (99%)
Class-RAG: Content Moderation with Retrieval Augmented Generation. (75%)
Attack as Defense: Run-time Backdoor Implantation for Image Content Protection. (61%)
Feint and Attack: Attention-Based Strategies for Jailbreaking and Protecting LLMs. (13%)
Stochastic Gradient Descent Jittering for Inverse Problems: Alleviating the Accuracy-Robustness Tradeoff. (10%)
Unlearning Backdoor Attacks for LLMs with Weak-to-Strong Knowledge Distillation. (5%)
Real-time Fake News from Adversarial Feedback. (3%)
Adversarial Score identity Distillation: Rapidly Surpassing the Teacher in One Step. (1%)
2024-10-17
MMAD-Purify: A Precision-Optimized Framework for Efficient and Scalable Multi-Modal Attacks. (99%)
DMGNN: Detecting and Mitigating Backdoor Attacks in Graph Neural Networks. (95%)
Adversarial Inception for Bounded Backdoor Poisoning in Deep Reinforcement Learning. (67%)
SPIN: Self-Supervised Prompt INjection. (67%)
Jailbreaking LLM-Controlled Robots. (56%)
Persistent Pre-Training Poisoning of LLMs. (33%)
Trojan Prompt Attacks on Graph Neural Networks. (4%)
Do LLMs Have Political Correctness? Analyzing Ethical Biases and Jailbreak Vulnerabilities in AI Systems. (2%)
2024-10-16
Golyadkin's Torment: Doppelg\"angers and Adversarial Vulnerability. (99%)
DAT: Improving Adversarial Robustness via Generative Amplitude Mix-up in Frequency Domain. (99%)
Boosting Imperceptibility of Stable Diffusion-based Adversarial Examples Generation with Momentum. (99%)
New Paradigm of Adversarial Training: Breaking Inherent Trade-Off between Accuracy and Robustness via Dummy Classes. (98%)
Perseus: Leveraging Common Data Patterns with Curriculum Learning for More Robust Graph Neural Networks. (92%)
Low-Rank Adversarial PGD Attack. (84%)
Data Defenses Against Large Language Models. (76%)
Hiding-in-Plain-Sight (HiPS) Attack on CLIP for Targetted Object Removal from Images. (61%)
NSmark: Null Space Based Black-box Watermarking Defense Framework for Pre-trained Language Models. (16%)
Reconstruction of Differentially Private Text Sanitization via Large Language Models. (4%)
Unitary Multi-Margin BERT for Robust Natural Language Processing. (4%)
Mitigating the Backdoor Effect for Multi-Task Model Merging via Safety-Aware Subspace. (2%)
FedGTST: Boosting Global Transferability of Federated Models via Statistics Tuning. (2%)
Consistency Calibration: Improving Uncertainty Calibration via Consistency among Perturbed Neighbors. (2%)
Efficient Optimization Algorithms for Linear Adversarial Training. (1%)
PromptExp: Multi-granularity Prompt Explanation of Large Language Models. (1%)
Long-Tailed Backdoor Attack Using Dynamic Data Augmentation Operations. (1%)
2024-10-15
Taking off the Rose-Tinted Glasses: A Critical Look at Adversarial ML Through the Lens of Evasion Attacks. (99%)
Efficient and Effective Universal Adversarial Attack against Vision-Language Pre-training Models. (98%)
Deciphering the Chaos: Enhancing Jailbreak Attacks via Adversarial Prompt Translation. (83%)
BeniFul: Backdoor Defense via Middle Feature Analysis for Deep Neural Networks. (82%)
Cognitive Overload Attack:Prompt Injection for Long Context. (62%)
AdvBDGen: Adversarially Fortified Prompt-Specific Fuzzy Backdoor Generator Against LLM Alignment. (31%)
Backdoor Attack on Vertical Federated Graph Neural Network Learning. (10%)
DiffGAN: A Test Generation Approach for Differential Testing of Deep Neural Networks. (10%)
Multi-round jailbreak attack on large language models. (4%)
Geometric Inductive Biases of Deep Networks: The Role of Data and Architecture. (3%)
G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks. (2%)
2024-10-14
Denial-of-Service Poisoning Attacks against Large Language Models. (92%)
Towards Calibrated Losses for Adversarial Robust Reject Option Classification. (86%)
Adversarially Robust Out-of-Distribution Detection Using Lyapunov-Stabilized Embeddings. (86%)
Adversarially Guided Stateful Defense Against Backdoor Attacks in Federated Deep Learning. (81%)
Feature Averaging: An Implicit Bias of Gradient Descent Leading to Non-Robustness in Neural Networks. (68%)
ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object Detection. (67%)
Generalized Adversarial Code-Suggestions: Exploiting Contexts of LLM-based Code-Completion. (15%)
Enhancing Robustness in Deep Reinforcement Learning: A Lyapunov Exponent Approach. (13%)
How to Backdoor Consistency Models? (12%)
The Implicit Bias of Structured State Space Models Can Be Poisoned With Clean Labels. (2%)
Context-Parametric Inversion: Why Instruction Finetuning May Not Actually Improve Context Reliance. (1%)
On Calibration of LLM-based Guard Models for Reliable Content Moderation. (1%)
Regularized Robustly Reliable Learners and Instance Targeted Attacks. (1%)
Automatically Generating Visual Hallucination Test Cases for Multimodal Large Language Models. (1%)
2024-10-13
S$^4$ST: A Strong, Self-transferable, faSt, and Simple Scale Transformation for Transferable Targeted Attack. (99%)
Understanding Robustness of Parameter-Efficient Tuning for Image Classification. (98%)
Out-of-Bounding-Box Triggers: A Stealthy Approach to Cheat Object Detectors. (75%)
Uncovering, Explaining, and Mitigating the Superficial Safety of Backdoor Defense. (67%)
BlackDAN: A Black-Box Multi-Objective Approach for Effective and Contextual Jailbreaking of Large Language Models. (13%)
Targeted Vaccine: Safety Alignment for Large Language Models against Harmful Fine-Tuning via Layer-wise Perturbation. (1%)
2024-10-12
Unlearn and Burn: Adversarial Machine Unlearning Requests Destroy Model Accuracy. (91%)
Robust 3D Point Clouds Classification based on Declarative Defenders. (2%)
2024-10-11
On the Adversarial Transferability of Generalized "Skip Connections". (99%)
Natural Language Induced Adversarial Images. (99%)
Fragile Giants: Understanding the Susceptibility of Models to Subpopulation Attacks. (70%)
AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents. (69%)
Training on Fake Labels: Mitigating Label Leakage in Split Learning via Secure Dimension Transformation. (62%)
PoisonBench: Assessing Large Language Model Vulnerability to Data Poisoning. (31%)
AttnGCG: Enhancing Jailbreaking Attacks on LLMs with Attention Manipulation. (31%)
The Good, the Bad and the Ugly: Watermarks, Transferable Attacks and Adversarial Defenses. (16%)
Impeding LLM-assisted Cheating in Introductory Programming Assignments via Adversarial Perturbation. (4%)
F2A: An Innovative Approach for Prompt Injection by Utilizing Feign Security Detection Agents. (1%)
RePD: Defending Jailbreak Attack through a Retrieval-based Prompt Decomposition Process. (1%)
JAILJUDGE: A Comprehensive Jailbreak Judge Benchmark with Multi-Agent Enhanced Explanation Evaluation Framework. (1%)
2024-10-10
Adversarial Training Can Provably Improve Robustness: Theoretical Analysis of Feature Learning Process Under Structured Data. (99%)
Time Traveling to Defend Against Adversarial Example Attacks in Image Classification. (99%)
Understanding Adversarially Robust Generalization via Weight-Curvature Index. (98%)
Invisibility Cloak: Disappearance under Human Pose Estimation via Backdoor Attacks. (92%)
A Survey on Physical Adversarial Attacks against Face Recognition Systems. (91%)
Towards Assurance of LLM Adversarial Robustness using Ontology-Driven Argumentation. (74%)
Bilinear MLPs enable weight-based mechanistic interpretability. (70%)
Adversarial Robustness Overestimation and Instability in TRADES. (67%)
RAB$^2$-DEF: Dynamic and explainable defense against adversarial attacks in Federated Learning to fair poor clients. (61%)
Poison-splat: Computation Cost Attack on 3D Gaussian Splatting. (10%)
A Closer Look at Machine Unlearning for Large Language Models. (4%)
2024-10-09
Break the Visual Perception: Adversarial Attacks Targeting Encoded Visual Tokens of Large Vision-Language Models. (99%)
Understanding Model Ensemble in Transferable Adversarial Attack. (99%)
Secure Video Quality Assessment Resisting Adversarial Attacks. (75%)
Can DeepFake Speech be Reliably Detected? (62%)
Data Taggants: Dataset Ownership Verification via Harmless Targeted Data Poisoning. (15%)
Average Certified Radius is a Poor Metric for Randomized Smoothing. (11%)
JPEG Inspired Deep Learning. (11%)
Adversarial Vulnerability as a Consequence of On-Manifold Inseparibility. (2%)
Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates. (2%)
Mind Your Questions! Towards Backdoor Attacks on Text-to-Visualization Models. (2%)
Utilize the Flow before Stepping into the Same River Twice: Certainty Represented Knowledge Flow for Refusal-Aware Instruction Tuning. (1%)
AdaRC: Mitigating Graph Structure Shifts during Test-Time. (1%)
PII-Scope: A Benchmark for Training Data PII Leakage Assessment in LLMs. (1%)
2024-10-08
Hyper Adversarial Tuning for Boosting Adversarial Robustness of Pretrained Large Vision Models. (99%)
DiffusionGuard: A Robust Defense Against Malicious Diffusion-based Image Editing. (98%)
Filtered Randomized Smoothing: A New Defense for Robust Modulation Classification. (98%)
CALoR: Towards Comprehensive Model Inversion Defense. (76%)
Polynomial Time Cryptanalytic Extraction of Deep Neural Networks in the Hard-Label Setting. (74%)
Training-free LLM-generated Text Detection by Mining Token Probability Sequences. (26%)
PFAttack: Stealthy Attack Bypassing Group Fairness in Federated Learning. (10%)
Recent advancements in LLM Red-Teaming: Techniques, Defenses, and Ethical Considerations. (10%)
2024-10-07
TaeBench: Improving Quality of Toxic Adversarial Examples. (99%)
AnyAttack: Towards Large-scale Self-supervised Generation of Targeted Adversarial Examples for Vision-Language Models. (99%)
LOTOS: Layer-wise Orthogonalization for Training Robust Ensembles. (99%)
Patch is Enough: Naturalistic Adversarial Patch against Vision-Language Pre-training Models. (95%)
MIBench: A Comprehensive Benchmark for Model Inversion Attack and Defense. (86%)
STOP! Camera Spoofing via the in-Vehicle IP Network. (83%)
Double Oracle Neural Architecture Search for Game Theoretic Deep Learning Models. (76%)
Collaboration! Towards Robust Neural Methods for Routing Problems. (70%)
Aligning LLMs to Be Robust Against Prompt Injection. (47%)
CAT: Concept-level backdoor ATtacks for Concept Bottleneck Models. (11%)
Defense-as-a-Service: Black-box Shielding against Backdoored Graph Models. (8%)
Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation. (1%)
2024-10-06
Suspiciousness of Adversarial Texts to Human. (99%)
On the Adversarial Risk of Test Time Adaptation: An Investigation into Realistic Test-Time Data Poisoning. (99%)
TA3: Testing Against Adversarial Attacks on Machine Learning Models. (67%)
Robustness Reprogramming for Representation Learning. (56%)
Towards Understanding and Enhancing Security of Proof-of-Training for DNN Model Ownership Verification. (2%)
Federated Learning Nodes Can Reconstruct Peers' Image Data. (1%)
2024-10-05
Harnessing Task Overload for Scalable Jailbreak Attacks on Large Language Models. (38%)
ConDa: Fast Federated Unlearning with Contribution Dampening. (1%)
2024-10-04
Mitigating Adversarial Perturbations for Deep Reinforcement Learning via Vector Quantization. (98%)
RAFT: Realistic Attacks to Fool Text Detectors. (96%)
A Brain-Inspired Regularizer for Adversarial Robustness. (92%)
Gradient-based Jailbreak Images for Multimodal Fusion Models. (16%)
You Know What I'm Saying -- Jailbreak Attack via Implicit Reference. (16%)
Impact of Regularization on Calibration and Robustness: from the Representation Space Perspective. (13%)
Make Interval Bound Propagation great again. (9%)
Classification-Denoising Networks. (9%)
Knowledge-Augmented Reasoning for EUAIA Compliance and Adversarial Robustness of LLMs. (2%)
BN-SCAFFOLD: controlling the drift of Batch Normalization statistics in Federated Learning. (1%)
Chain-of-Jailbreak Attack for Image Generation Models via Editing Step by Step. (1%)
2024-10-03
SCA: Highly Efficient Semantic-Consistent Unrestricted Adversarial Attack. (99%)
Towards Universal Certified Robustness with Multi-Norm Training. (26%)
Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents. (15%)
AutoDAN-Turbo: A Lifelong Agent for Strategy Self-Exploration to Jailbreak LLMs. (11%)
Demonstration Attack against In-Context Learning for Code Intelligence. (10%)
Unveiling AI's Blind Spots: An Oracle for In-Domain, Out-of-Domain, and Adversarial Errors. (3%)
Jailbreak Antidote: Runtime Safety-Utility Balance via Sparse Representation Adjustment in Large Language Models. (3%)
MTDNS: Moving Target Defense for Resilient DNS Infrastructure. (2%)
Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems. (1%)
IndicSentEval: How Effectively do Multilingual Transformer Models encode Linguistic Properties for Indic Languages? (1%)
BACKTIME: Backdoor Attacks on Multivariate Time Series Forecasting. (1%)
Optimizing Adaptive Attacks against Content Watermarks for Language Models. (1%)
Universally Optimal Watermarking Schemes for LLMs: from Theory to Practice. (1%)
Buckle Up: Robustifying LLMs at Every Customization Stage via Data Curation. (1%)
2024-10-02
On Using Certified Training towards Empirical Robustness. (99%)
Impact of White-Box Adversarial Attacks on Convolutional Neural Networks. (99%)
Signal Adversarial Examples Generation for Signal Detection Network via White-Box Attack. (99%)
MOREL: Enhancing Adversarial Robustness through Multi-Objective Representation Learning. (99%)
Fake It Until You Break It: On the Adversarial Robustness of AI-generated Image Detectors. (98%)
"No Matter What You Do": Purifying GNN Models via Backdoor Unlearning. (93%)
Social Media Authentication and Combating Deepfakes using Semi-fragile Invisible Image Watermarking. (82%)
The Unlikely Hero: Nonideality in Analog Photonic Neural Networks as Built-in Defender Against Adversarial Attacks. (76%)
BadCM: Invisible Backdoor Attack Against Cross-Modal Learning. (13%)
Controlled Generation of Natural Adversarial Documents for Stealthy Retrieval Poisoning. (13%)
Automated Red Teaming with GOAT: the Generative Offensive Agent Tester. (11%)
Information-Theoretical Principled Trade-off between Jailbreakability and Stealthiness on Vision Language Models. (8%)
One Wave to Explain Them All: A Unifying Perspective on Post-hoc Explainability. (1%)
2024-10-01
Empirical Perturbation Analysis of Linear System Solvers from a Data Poisoning Perspective. (54%)
Adversarial Suffixes May Be Features Too! (15%)
2024-09-30
Characterizing Model Robustness via Natural Input Gradients. (92%)
Robust LLM safeguarding via refusal feature adversarial training. (80%)
Resonance Reduction Against Adversarial Attacks in Dynamic Networks via Eigenspectrum Optimization. (76%)
Navigating Threats: A Survey of Physical Adversarial Attacks on LiDAR Perception Systems in Autonomous Vehicles. (45%)
VLMGuard: Defending VLMs against Malicious Prompts via Unlabeled Data. (8%)
2024-09-29
MASKDROID: Robust Android Malware Detection with Masked Graph Representations. (99%)
Adversarial Examples for DNA Classification. (98%)
Discerning the Chaos: Detecting Adversarial Perturbations while Disentangling Intentional from Unintentional Noises. (86%)
BadHMP: Backdoor Attack against Human Motion Prediction. (61%)
Nonideality-aware training makes memristive networks more robust to adversarial attacks. (38%)
Infighting in the Dark: Multi-Labels Backdoor Attack in Federated Learning. (33%)
Towards Robust Extractive Question Answering Models: Rethinking the Training Methodology. (10%)
Learning Robust Policies via Interpretable Hamilton-Jacobi Reachability-Guided Disturbances. (5%)
IDEAW: Robust Neural Audio Watermarking with Invertible Dual-Embedding. (1%)
Can Models Learn Skill Composition from Examples? (1%)
2024-09-28
Efficient Backdoor Defense in Multimodal Contrastive Learning: A Token-Level Unlearning Method for Mitigating Threats. (74%)
GenTel-Safe: A Unified Benchmark and Shielding Framework for Defending Against Prompt Injection Attacks. (13%)
Leveraging MTD to Mitigate Poisoning Attacks in Decentralized FL with Non-IID Data. (11%)
Survey of Security and Data Attacks on Machine Unlearning In Financial and E-Commerce. (2%)
Privacy Attack in Federated Learning is Not Easy: An Experimental Study. (1%)
2024-09-27
Adversarial Challenges in Network Intrusion Detection Systems: Research Insights and Future Prospects. (96%)
Enhancing Robustness of Graph Neural Networks through p-Laplacian. (12%)
Efficient Noise Mitigation for Enhancing Inference Accuracy in DNNs on Mixed-Signal Accelerators. (1%)
In-depth Analysis of Privacy Threats in Federated Learning for Medical Data. (1%)
2024-09-26
Showing Many Labels in Multi-label Classification Models: An Empirical Study of Adversarial Examples. (98%)
Cross-Modality Attack Boosted by Gradient-Evolutionary Multiform Optimization. (98%)
Discovering New Shadow Patterns for Black-Box Attacks on Lane Detection of Autonomous Vehicles. (97%)
Improving Fast Adversarial Training via Self-Knowledge Guidance. (82%)
Faithfulness and the Notion of Adversarial Sensitivity in NLP Explanations. (69%)
CleanerCLIP: Fine-grained Counterfactual Semantic Augmentation for Backdoor Defense in Contrastive Learning. (69%)
DarkSAM: Fooling Segment Anything Model to Segment Nothing. (68%)
Perturb, Attend, Detect and Localize (PADL): Robust Proactive Image Defense. (56%)
Development of an Edge Resilient ML Ensemble to Tolerate ICS Adversarial Attacks. (54%)
Backdoor Attacks for LLMs with Weak-To-Strong Knowledge Distillation. (15%)
Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey. (15%)
Dark Miner: Defend against unsafe generation for text-to-image diffusion models. (5%)
An Adversarial Perspective on Machine Unlearning for AI Safety. (2%)
Revolutionizing Payload Inspection: A Self-Supervised Journey to Precision with Few Shots. (2%)
2024-09-25
Improving the Shortest Plank: Vulnerability-Aware Adversarial Training for Robust Recommender System. (93%)
A Hybrid Quantum-Classical AI-Based Detection Strategy for Generative Adversarial Network-Based Deepfake Attacks on an Autonomous Vehicle Traffic Sign Classification System. (82%)
RED QUEEN: Safeguarding Large Language Models against Concealed Multi-Turn Jailbreaking. (75%)
Transient Adversarial 3D Projection Attacks on Object Detection in Autonomous Driving. (67%)
Examining the Rat in the Tunnel: Interpretable Multi-Label Classification of Tor-based Malware. (45%)
SWE2: SubWord Enriched and Significant Word Emphasized Framework for Hate Speech Detection. (38%)
SHEATH: Defending Horizontal Collaboration for Distributed CNNs against Adversarial Noise. (22%)
Claim-Guided Textual Backdoor Attack for Practical Applications. (10%)
Cat-and-Mouse Satellite Dynamics: Divergent Adversarial Reinforcement Learning for Contested Multi-Agent Space Operations. (1%)
2024-09-24
Adversarial Backdoor Defense in CLIP. (99%)
Revisiting Acoustic Features for Robust ASR. (84%)
Adversarial Watermarking for Face Recognition. (80%)
Proactive Schemes: A Survey of Adversarial Attacks for Social Good. (54%)
Privacy Evaluation Benchmarks for NLP Models. (45%)
Towards Robust Object Detection: Identifying and Removing Backdoors via Module Inconsistency Analysis. (33%)
PACE: Poisoning Attacks on Learned Cardinality Estimation. (4%)
2024-09-23
Improving Adversarial Robustness for 3D Point Cloud Recognition at Test-Time through Purified Self-Training. (96%)
Interpretability-Guided Test-Time Adversarial Defense. (87%)
Effective and Evasive Fuzz Testing-Driven Jailbreaking Attacks against LLMs. (87%)
Data Poisoning-based Backdoor Attack Framework against Supervised Learning Rules of Spiking Neural Networks. (68%)
Attack Atlas: A Practitioner's Perspective on Challenges and Pitfalls in Red Teaming GenAI. (47%)
PROMPTFUZZ: Harnessing Fuzzing Techniques for Robust Testing of Prompt Injection in LLMs. (33%)
Log-normal Mutations and their Use in Detecting Surreptitious Fake Images. (13%)
Curb Your Attention: Causal Attention Gating for Robust Trajectory Prediction in Autonomous Driving. (12%)
Toward Mixture-of-Experts Enabled Trustworthy Semantic Communication for 6G Networks. (5%)
Room Impulse Responses help attackers to evade Deep Fake Detection. (1%)
AIM 2024 Sparse Neural Rendering Challenge: Dataset and Benchmark. (1%)
UTrace: Poisoning Forensics for Private Collaborative Learning. (1%)
SDBA: A Stealthy and Long-Lasting Durable Backdoor Attack in Federated Learning. (1%)
2024-09-22
Enhancing LLM-based Autonomous Driving Agents to Mitigate Perception Attacks. (10%)
Evaluating the Performance and Robustness of LLMs in Materials Science Q&A and Property Predictions. (1%)
2024-09-21
Cloud Adversarial Example Generation for Remote Sensing Image Classification. (99%)
Adversarial Attacks on Parts of Speech: An Empirical Study in Text-to-Image Generation. (98%)
When Witnesses Defend: A Witness Graph Topological Layer for Adversarial Graph Learning. (69%)
PathSeeker: Exploring LLM Security Vulnerabilities with a Reinforcement Learning-Based Jailbreak Approach. (62%)
ESPERANTO: Evaluating Synthesized Phrases to Enhance Robustness in AI Detection for Text Origination. (10%)
Perfect Gradient Inversion in Federated Learning: A New Paradigm from the Hidden Subset Sum Problem. (8%)
Data-centric NLP Backdoor Defense from the Lens of Memorization. (4%)
2024-09-20
Efficient Visualization of Neural Networks with Generative Models and Adversarial Perturbations. (99%)
ViTGuard: Attention-aware Detection against Adversarial Examples for Vision Transformer. (99%)
Certified Adversarial Robustness via Partition-based Randomized Smoothing. (81%)
ID-Guard: A Universal Framework for Combating Facial Manipulation via Breaking Identification. (76%)
Persistent Backdoor Attacks in Continual Learning. (73%)
Relationship between Uncertainty in DNNs and Adversarial Attacks. (70%)
PureDiffusion: Using Backdoor to Counter Backdoor in Generative Diffusion Models. (61%)
On the Feasibility of Fully AI-automated Vishing Attacks. (1%)
2024-09-19
Deep generative models as an adversarial attack strategy for tabular machine learning. (99%)
TEAM: Temporal Adversarial Examples Attack Model against Network Intrusion Detection System Applied to RNN. (99%)
Hidden Activations Are Not Enough: A General Approach to Neural Network Predictions. (98%)
Defending against Reverse Preference Attacks is Difficult. (83%)
Revisiting Semi-supervised Adversarial Robustness via Noise-aware Online Robust Distillation. (45%)
VCAT: Vulnerability-aware and Curiosity-driven Adversarial Training for Enhancing Autonomous Vehicle Robustness. (26%)
Data Poisoning and Leakage Analysis in Federated Learning. (11%)
Manipulation Facing Threats: Evaluating Physical Vulnerabilities in End-to-End Vision Language Action Models. (2%)
Hidden in Plain Sound: Environmental Backdoor Poisoning Attacks on Whisper, and Mitigations. (2%)
2024-09-18
Enhancing 3D Robotic Vision Robustness by Minimizing Adversarial Mutual Information through a Curriculum Training Approach. (99%)
ITPatch: An Invisible and Triggered Physical Adversarial Patch against Traffic Sign Recognition. (99%)
NPAT Null-Space Projected Adversarial Training Towards Zero Deterioration. (96%)
LLM-Powered Text Simulation Attack Against ID-Free Recommender Systems. (76%)
PAD-FT: A Lightweight Defense for Backdoor Attacks via Data Purification and Fine-Tuning. (68%)
A constrained optimization approach to improve robustness of neural networks. (54%)
Understanding Implosion in Text-to-Image Generative Models. (2%)
2024-09-17
Golden Ratio Search: A Low-Power Adversarial Attack for Deep Learning based Modulation Classification. (98%)
EIA: Environmental Injection Attack on Generalist Web Agents for Privacy Leakage. (76%)
Contextual Breach: Assessing the Robustness of Transformer-based QA Models. (56%)
Hard-Label Cryptanalytic Extraction of Neural Network Models. (2%)
2024-09-16
Towards Physically-Realizable Adversarial Attacks in Embodied Vision Navigation. (82%)
CaBaGe: Data-Free Model Extraction using ClAss BAlanced Generator Ensemble. (2%)
Realistic Extreme Behavior Generation for Improved AV Testing. (1%)
Jailbreaking Large Language Models with Symbolic Mathematics. (1%)
Speaker Contrastive Learning for Source Speaker Tracing. (1%)
2024-09-15
Revisiting Physical-World Adversarial Attack on Traffic Sign Recognition: A Commercial Systems Perspective. (98%)
Federated Learning in Adversarial Environments: Testbed Design and Poisoning Resilience in Cybersecurity. (8%)
2024-09-14
Real-world Adversarial Defense against Patch Attacks based on Diffusion Model. (99%)
2024-09-13
XSub: Explanation-Driven Adversarial Attack against Blackbox Classifiers via Feature Substitution. (95%)
Are Existing Road Design Guidelines Suitable for Autonomous Vehicles? (41%)
Clean Label Attacks against SLU Systems. (31%)
FAST: Boosting Uncertainty-based Test Prioritization Methods for Neural Networks via Feature Selection. (15%)
2024-09-12
LoRID: Low-Rank Iterative Diffusion for Adversarial Purification. (99%)
High-Frequency Anti-DreamBooth: Robust Defense against Personalized Image Synthesis. (93%)
FedProphet: Memory-Efficient Federated Adversarial Training via Theoretic-Robustness and Low-Inconsistency Cascade Learning. (92%)
Exploiting Supervised Poison Vulnerability to Strengthen Self-Supervised Defense. (73%)
Sub-graph Based Diffusion Model for Link Prediction. (9%)
Unleashing Worms and Extracting Data: Escalating the Outcome of Attacks against RAG-based Inference in Scale and Severity Using Jailbreaking. (1%)
Risks When Sharing LoRA Fine-Tuned Diffusion Model Weights. (1%)
2024-09-11
Module-wise Adaptive Adversarial Training for End-to-end Autonomous Driving. (99%)
Securing Vision-Language Models with a Robust Encoder Against Jailbreak and Adversarial Attacks. (98%)
Introducing Perturb-ability Score (PS) to Enhance Robustness Against Evasion Adversarial Attacks on ML-NIDS. (97%)
D-CAPTCHA++: A Study of Resilience of Deepfake CAPTCHA under Transferable Imperceptible Adversarial Attack. (93%)
A Cost-Aware Approach to Adversarial Robustness in Neural Networks. (84%)
Attack End-to-End Autonomous Driving through Module-Wise Noise. (74%)
On the Vulnerability of Applying Retrieval-Augmented Generation within Knowledge-Intensive Application Domains. (67%)
Enhancing adversarial robustness in Natural Language Inference using explanations. (67%)
AdvLogo: Adversarial Patch Attack against Object Detectors based on Diffusion Models. (64%)
Understanding Knowledge Drift in LLMs through Misinformation. (1%)
2024-09-10
Unrevealed Threats: A Comprehensive Study of the Adversarial Robustness of Underwater Image Enhancement Models. (99%)
Advancing Hybrid Defense for Byzantine Attacks in Federated Learning. (84%)
Adversarial Attacks to Multi-Modal Models. (76%)
DV-FSR: A Dual-View Target Attack Framework for Federated Sequential Recommendation. (67%)
2024-09-09
Seeing Through the Mask: Rethinking Adversarial Examples for CAPTCHAs. (99%)
Adversarial Attacks on Data Attribution. (99%)
Input Space Mode Connectivity in Deep Neural Networks. (83%)
Unlearning or Concealment? A Critical Analysis and Evaluation Metrics for Unlearning in Diffusion Models. (83%)
On the Weaknesses of Backdoor-based Model Watermarking: An Information-theoretic Perspective. (33%)
2024-09-08
PIP: Detecting Adversarial Examples in Large Vision-Language Models via Attention Patterns of Irrelevant Probe Questions. (99%)
2DSig-Detect: a semi-supervised framework for anomaly detection on image data using 2D-signatures. (87%)
Vision-fused Attack: Advancing Aggressive and Stealthy Adversarial Text against Neural Machine Translation. (67%)
Natias: Neuron Attribution based Transferable Image Adversarial Steganography. (67%)
2024-09-07
Phrase-Level Adversarial Training for Mitigating Bias in Neural Network-based Automatic Essay Scoring. (86%)
PIXHELL Attack: Leaking Sensitive Information from Air-Gap Computers via `Singing Pixels'. (80%)
Top-GAP: Integrating Size Priors in CNNs for more Interpretability, Robustness, and Bias Mitigation. (12%)
2024-09-06
Learning to Learn Transferable Generative Attack for Person Re-Identification. (99%)
PANTS: Practical Adversarial Network Traffic Samples against ML-powered Networking Classifiers. (99%)
Secure Traffic Sign Recognition: An Attention-Enabled Universal Image Inpainting Mechanism against Light Patch Attacks. (83%)
Mind The Gap: Can Air-Gaps Keep Your Private Data Secure? (74%)
Exploiting the Data Gap: Utilizing Non-ignorable Missingness to Manipulate Model Learning. (38%)
Context is the Key: Backdoor Attacks for In-Context Learning with Vision Transformers. (8%)
Dual-stream Feature Augmentation for Domain Generalization. (8%)
2024-09-05
A practical approach to evaluating the adversarial distance for machine learning classifiers. (98%)
Non-Uniform Illumination Attack for Fooling Convolutional Neural Networks. (92%)
Limited but consistent gains in adversarial robustness by co-training object recognition models with human EEG. (31%)
Recent Advances in Attack and Defense Approaches of Large Language Models. (4%)
WaterMAS: Sharpness-Aware Maximization for Neural Network Watermarking. (3%)
Understanding Data Importance in Machine Learning Attacks: Does Valuable Data Pose Greater Harm? (1%)
2024-09-04
Bypassing DARCY Defense: Indistinguishable Universal Adversarial Triggers. (99%)
OpenFact at CheckThat! 2024: Combining Multiple Attack Methods for Effective Adversarial Text Generation. (99%)
TASAR: Transferable Attack on Skeletal Action Recognition. (92%)
Adversarial Attacks on Machine Learning-Aided Visualizations. (83%)
Transfer-based Adversarial Poisoning Attacks for Online (MIMO-)Deep Receviers. (76%)
Boosting Certificate Robustness for Time Series Classification with Efficient Self-Ensemble. (70%)
AdvSecureNet: A Python Toolkit for Adversarial Machine Learning. (33%)
Active Fake: DeepFake Camouflage. (13%)
Well, that escalated quickly: The Single-Turn Crescendo Attack (STCA). (2%)
2024-09-03
Exploiting the Vulnerability of Large Language Models via Defense-Aware Architectural Backdoor. (97%)
Dynamic Guidance Adversarial Distillation with Enhanced Teacher Knowledge. (92%)
NoiseAttack: An Evasive Sample-Specific Multi-Targeted Backdoor Attack Through White Gaussian Noise. (16%)
Reassessing Noise Augmentation Methods in the Context of Adversarial Speech. (5%)
On the Vulnerability of Skip Connections to Model Inversion Attacks. (3%)
2024-09-02
One-Index Vector Quantization Based Adversarial Attack on Image Classification. (99%)
Adversarial Pruning: A Survey and Benchmark of Pruning Methods for Adversarial Robustness. (99%)
Phantom: Untargeted Poisoning Attacks on Semi-Supervised Learning (Full Version). (68%)
Defending against Model Inversion Attacks via Random Erasing. (64%)
CLIBE: Detecting Dynamic Backdoors in Transformer-based NLP Models. (62%)
Purification-Agnostic Proxy Learning for Agentic Copyright Watermarking against Adversarial Evidence Forgery. (26%)
Unveiling the Vulnerability of Private Fine-Tuning in Split-Based Frameworks for Large Language Models: A Bidirectionally Enhanced Attack. (26%)
A Review of Image Retrieval Techniques: Data Augmentation and Adversarial Learning Approaches. (16%)
Spatial-Aware Conformal Prediction for Trustworthy Hyperspectral Image Classification. (1%)
2024-09-01
Comprehensive Botnet Detection by Mitigating Adversarial Attacks, Navigating the Subtleties of Perturbation Distances and Fortifying Predictions with Conformal Layers. (99%)
Accurate Forgetting for All-in-One Image Restoration Model. (83%)
The Dark Side of Human Feedback: Poisoning Large Language Models via User Inputs. (26%)
Fisher Information guided Purification against Backdoor Attacks. (12%)
2024-08-31
HSF: Defending against Jailbreak Attacks with Hidden State Filtering. (75%)
Is Difficulty Calibration All We Need? Towards More Practical Membership Inference Attacks. (15%)
Robust off-policy Reinforcement Learning via Soft Constrained Adversary. (4%)
2024-08-30
LightPure: Realtime Adversarial Image Purification for Mobile Devices Using Diffusion Models. (92%)
Instant Adversarial Purification with Adversarial Consistency Distillation. (33%)
PRADA: Proactive Risk Assessment and Mitigation of Misinformed Demand Attacks on Navigational Route Recommendations. (8%)
Evaluating Reliability in Medical DNNs: A Critical Analysis of Feature and Confidence-Based OOD Detection. (1%)
2024-08-29
PromptSmooth: Certifying Robustness of Medical Vision-Language Models via Prompt Learning. (92%)
STEREO: Towards Adversarially Robust Concept Erasing from Text-to-Image Generation Models. (83%)
SFR-GNN: Simple and Fast Robust GNNs against Structural Attacks. (67%)
Analyzing Inference Privacy Risks Through Gradients in Machine Learning. (54%)
Tex-ViT: A Generalizable, Robust, Texture-based dual-branch cross-attention deepfake detector. (12%)
2024-08-28
Evaluating Model Robustness Using Adaptive Sparse L0 Regularization. (99%)
Network transferability of adversarial patches in real-time object detection. (83%)
Defending Text-to-image Diffusion Models: Surprising Efficacy of Textual Perturbations Against Backdoor Attacks. (83%)
Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation. (47%)
VFLIP: A Backdoor Defense for Vertical Federated Learning via Identification and Purification. (2%)
FRACTURED-SORRY-Bench: Framework for Revealing Attacks in Conversational Turns Undermining Refusal Efficacy and Defenses over SORRY-Bench (Automated Multi-shot Jailbreaks). (1%)
2024-08-27
Adversarial Attacks and Defenses in Multivariate Time-Series Forecasting for Smart and Connected Infrastructures. (99%)
Certified Causal Defense with Generalizable Robustness. (99%)
Improving Adversarial Robustness in Android Malware Detection by Reducing the Impact of Spurious Correlations. (99%)
Adversarial Manhole: Challenging Monocular Depth Estimation and Semantic Segmentation Models with Patch Attack. (98%)
LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet. (12%)
Investigating Coverage Criteria in Large Language Models: An In-Depth Study Through Jailbreak Attacks. (11%)
Detecting AI Flaws: Target-Driven Attacks on Internal Faults in Language Models. (8%)
SpecGuard: Specification Aware Recovery for Robotic Autonomous Vehicles from Physical Attacks. (3%)
EmoAttack: Utilizing Emotional Voice Conversion for Speech Backdoor Attacks on Deep Speech Classification Models. (2%)
2024-08-26
TART: Boosting Clean Accuracy Through Tangent Direction Guided Adversarial Training. (99%)
2D-Malafide: Adversarial Attacks Against Face Deepfake Detection Systems. (99%)
Feedback-based Modal Mutual Search for Attacking Vision-Language Pre-training Models. (99%)
Celtibero: Robust Layered Aggregation for Federated Learning. (92%)
Dual Adversarial Perturbators Generate rich Views for Recommendation. (5%)
Investigating the Effectiveness of Bayesian Spam Filters in Detecting LLM-modified Spam Mails. (1%)
Surprisingly Fragile: Assessing and Addressing Prompt Instability in Multimodal Foundation Models. (1%)
2024-08-25
On the Robustness of Kolmogorov-Arnold Networks: An Adversarial Perspective. (98%)
RT-Attack: Jailbreaking Text-to-Image Models via Random Token. (97%)
TF-Attack: Transferable and Fast Adversarial Attacks on Large Language Models. (96%)
Generalization of Graph Neural Networks is Robust to Model Mismatch. (1%)
2024-08-24
Probing the Robustness of Vision-Language Pretrained Models: A Multimodal Adversarial Attack Approach. (99%)
Evaluating the Robustness of LiDAR-based 3D Obstacles Detection and Its Impacts on Autonomous Driving Systems. (1%)
2024-08-23
Dynamic Label Adversarial Training for Deep Learning Robustness Against Adversarial Attacks. (99%)
Toward Improving Synthetic Audio Spoofing Detection Robustness via Meta-Learning and Disentangled Training With Adversarial Examples. (98%)
Disentangled Training with Adversarial Examples For Robust Small-footprint Keyword Spotting. (83%)
Protecting against simultaneous data poisoning attacks. (54%)
2024-08-22
Enhancing Transferability of Adversarial Attacks with GE-AdvGAN+: A Comprehensive Framework for Gradient Editing. (99%)
Leveraging Information Consistency in Frequency and Spatial Domain for Adversarial Attacks. (99%)
MakeupAttack: Feature Space Black-box Backdoor Attack on Face Recognition via Makeup Transfer. (98%)
BankTweak: Adversarial Attack against Multi-Object Trackers by Manipulating Feature Banks. (80%)
On the Credibility of Backdoor Attacks Against Object Detectors in the Physical World. (75%)
Quantifying Psychological Sophistication of Malicious Emails. (2%)
Is Generative AI the Next Tactical Cyber Weapon For Threat Actors? Unforeseen Implications of AI Generated Cyber Attacks. (2%)
VALE: A Multimodal Visual and Language Explanation Framework for Image Classifiers using eXplainable AI and Language Models. (2%)
BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks on Large Language Models. (2%)
2024-08-21
Query-Efficient Video Adversarial Attack with Stylized Logo. (99%)
Pixel Is Not A Barrier: An Effective Evasion Attack for Pixel-Domain Diffusion Models. (74%)
A Practical Trigger-Free Backdoor Attack on Neural Networks. (67%)
First line of defense: A robust first layer mitigates adversarial attacks. (54%)
Exploring Robustness of Visual State Space model against Backdoor Attacks. (45%)
Against All Odds: Overcoming Typology, Script, and Language Confusion in Multilingual Embedding Inversion Attacks. (26%)
Latent Feature and Attention Dual Erasure Attack against Multi-View Diffusion Models for 3D Assets Protection. (12%)
Large Language Models are Good Attackers: Efficient and Stealthy Textual Backdoor Attacks. (10%)
2024-08-20
GAIM: Attacking Graph Neural Networks via Adversarial Influence Maximization. (99%)
Correlation Analysis of Adversarial Attack in Time Series Classification. (99%)
Privacy-preserving Universal Adversarial Defense for Black-box Models. (99%)
MsMemoryGAN: A Multi-scale Memory GAN for Palm-vein Adversarial Purification. (99%)
Revisiting Min-Max Optimization Problem in Adversarial Training. (97%)
Prompt-Agnostic Adversarial Perturbation for Customized Diffusion Models. (97%)
Iterative Window Mean Filter: Thwarting Diffusion-based Adversarial Purification. (87%)
Adversarial Attack for Explanation Robustness of Rationalization Models. (82%)
Towards Robust Knowledge Unlearning: An Adversarial Framework for Assessing and Improving Unlearning Robustness in Large Language Models. (73%)
A Grey-box Attack against Latent Diffusion Model-based Image Editing by Posterior Collapse. (68%)
Security Assessment of Hierarchical Federated Deep Learning. (67%)
Improving Out-of-Distribution Data Handling and Corruption Resistance via Modern Hopfield Networks. (54%)
Ferret: Faster and Effective Automated Red Teaming with Reward-Based Scoring Technique. (50%)
Makeup-Guided Facial Privacy Protection via Untrained Neural Network Priors. (33%)
EEG-Defender: Defending against Jailbreak through Early Exit Generation of Large Language Models. (31%)
Accelerating the Surrogate Retraining for Poisoning Attacks against Recommender Systems. (26%)
Unlocking Adversarial Suffix Optimization Without Affirmative Phrases: Efficient Black-box Jailbreaking via LLM as Optimizer. (10%)
Security Attacks on LLM-based Code Completion Tools. (8%)
MEGen: Generative Backdoor in Large Language Models via Model Editing. (2%)
Learning Randomized Algorithms with Transformers. (1%)
2024-08-19
Robust Image Classification: Defensive Strategies against FGSM and PGD Adversarial Attacks. (99%)
Detecting Adversarial Attacks in Semantic Segmentation via Uncertainty Estimation: A Deep Analysis. (99%)
Segment-Anything Models Achieve Zero-shot Robustness in Autonomous Driving. (98%)
Criticality Leveraged Adversarial Training (CLAT) for Boosted Performance via Parameter Efficiency. (31%)
The Brittleness of AI-Generated Image Watermarking Techniques: Examining Their Robustness Against Visual Paraphrasing Attacks. (5%)
Transferring Backdoors between Large Language Models by Knowledge Distillation. (2%)
Enhance Modality Robustness in Text-Centric Multimodal Alignment with Adversarial Prompting. (1%)
Perfectly Undetectable Reflection and Scaling False Data Injection Attacks via Affine Transformation on Mobile Robot Trajectory Tracking Control. (1%)
2024-08-18
Enhancing Adversarial Transferability with Adversarial Weight Tuning. (99%)
Regularization for Adversarial Robust Learning. (41%)
Adversarial Attacked Teacher for Unsupervised Domain Adaptive Object Detection. (31%)
GANPrompt: Enhancing Robustness in LLM-Based Recommendations with GAN-Enhanced Diversity Prompts. (1%)
Global BGP Attacks that Evade Route Monitoring. (1%)
2024-08-17
Attack Anything: Blind DNNs via Universal Background Adversarial Attack. (99%)
Training Verifiably Robust Agents Using Set-Based Reinforcement Learning. (75%)
DiffZOO: A Purely Query-Based Black-Box Attack for Red-teaming Text-to-Image Generative Model via Zeroth Order Optimization. (67%)
PADetBench: Towards Benchmarking Physical Attacks against Object Detection. (62%)
Malacopula: adversarial automatic speaker verification attacks using a neural-based generalised Hammerstein model. (31%)
BaThe: Defense against the Jailbreak Attack in Multimodal Large Language Models by Treating Harmful Instruction as Backdoor Trigger. (10%)
Characterizing and Evaluating the Reliability of LLMs against Jailbreak Attacks. (5%)
PREMAP: A Unifying PREiMage APproximation Framework for Neural Networks. (2%)
Out-of-distribution materials property prediction using adversarial learning based fine-tuning. (1%)
2024-08-16
Ask, Attend, Attack: A Effective Decision-Based Black-Box Targeted Attack for Image-to-Text Models. (98%)
Towards Physical World Backdoor Attacks against Skeleton Action Recognition. (93%)
LEVIS: Large Exact Verifiable Input Spaces for Neural Networks. (87%)
Visual-Friendly Concept Protection via Selective Adversarial Perturbations. (75%)
Can Large Language Models Improve the Adversarial Robustness of Graph Neural Networks? (75%)
Mitigating Backdoor Attacks in Federated Learning via Flipping Weight Updates of Low-Activation Input Neurons. (1%)
2024-08-15
DFT-Based Adversarial Attack Detection in MRI Brain Imaging: Enhancing Diagnostic Accuracy in Alzheimer's Case Studies. (99%)
A Multi-task Adversarial Attack Against Face Authentication. (98%)
Evaluating Text Classification Robustness to Part-of-Speech Adversarial Examples. (98%)
Unlearnable Examples Detection via Iterative Filtering. (88%)
A Survey of Trojan Attacks and Defenses to Deep Neural Networks. (78%)
Efficient Image-to-Image Diffusion Classifier for Adversarial Robustness. (76%)
Prefix Guidance: A Steering Wheel for Large Language Models to Defend Against Jailbreak Attacks. (74%)
$\textit{MMJ-Bench}$: A Comprehensive Study on Jailbreak Attacks and Defenses for Multimodal Large Language Models. (70%)
Random Gradient Masking as a Defensive Measure to Deep Leakage in Federated Learning. (8%)
A Robust Multi-Stage Intrusion Detection System for In-Vehicle Network Security using Hierarchical Federated Learning. (2%)
2024-08-14
Enhancing Adversarial Attacks via Parameter Adaptive Adversarial Attack. (99%)
TabularBench: Benchmarking Adversarial Robustness for Tabular Deep Learning in Real-world Use-cases. (98%)
Robust Active Learning (RoAL): Countering Dynamic Adversaries in Active Learning with Elastic Weight Consolidation. (80%)
Achieving Data Efficient Neural Networks with Hybrid Concept-based Models. (70%)
Sonic: Fast and Transferable Data Poisoning on Clustering Algorithms. (67%)
BadMerging: Backdoor Attacks Against Model Merging. (47%)
BAPLe: Backdoor Attacks on Medical Foundational Models using Prompt Learning. (38%)
Cognitive Networks and Performance Drive fMRI-Based State Classification Using DNN Models. (1%)
2024-08-13
DePatch: Towards Robust Adversarial Patch for Evading Person Detectors in the Real World. (92%)
Robust Black-box Testing of Deep Neural Networks using Co-Domain Coverage. (12%)
Imagen 3. (10%)
2024-08-12
Towards Adversarial Robustness via Debiased High-Confidence Logit Alignment. (99%)
Fooling SHAP with Output Shuffling Attacks. (81%)
Understanding Byzantine Robustness in Federated Learning with A Black-box Server. (13%)
2024-08-11
Improving Adversarial Transferability with Neighbourhood Gradient Information. (99%)
Classifier Guidance Enhances Diffusion-based Adversarial Purification by Preserving Predictive Information. (98%)
Kov: Transferable and Naturalistic Black-Box LLM Attacks using Markov Decision Processes and Tree Search. (9%)
2024-08-10
ReToMe-VA: Recursive Token Merging for Video Diffusion-based Unrestricted Adversarial Attack. (99%)
StealthDiffusion: Towards Evading Diffusion Forensic Detection through Diffusion Model. (99%)
PointNCBW: Towards Dataset Ownership Verification for Point Clouds via Negative Clean-label Backdoor Watermark. (13%)
2024-08-09
Modeling Electromagnetic Signal Injection Attacks on Camera-based Smart Systems: Applications and Mitigation. (84%)
A Jailbroken GenAI Model Can Cause Substantial Harm: GenAI-powered Applications are Vulnerable to PromptWares. (2%)
Rag and Roll: An End-to-End Evaluation of Indirect Prompt Manipulations in LLM-based Application Frameworks. (2%)
TrajFM: A Vehicle Trajectory Foundation Model for Region and Task Transferability. (1%)
2024-08-08
Constructing Adversarial Examples for Vertical Federated Learning: Optimal Client Corruption through Multi-Armed Bandit. (99%)
Adversarially Robust Industrial Anomaly Detection Through Diffusion Model. (99%)
Ensemble everything everywhere: Multi-scale aggregation for adversarial robustness. (99%)
Eliminating Backdoors in Neural Code Models via Trigger Inversion. (92%)
Improving Network Interpretability via Explanation Consistency Evaluation. (81%)
Unveiling Hidden Visual Information: A Reconstruction Attack Against Adversarial Visual Information Hiding. (80%)
Towards Resilient and Efficient LLMs: A Comparative Study of Efficiency, Performance, and Adversarial Robustness. (67%)
Stability Analysis of Equivariant Convolutional Representations Through The Lens of Equivariant Multi-layered CKNs. (61%)
h4rm3l: A Dynamic Benchmark of Composable Jailbreak Attacks for LLM Safety Assessment. (15%)
VideoQA in the Era of LLMs: An Empirical Study. (1%)
2024-08-07
Investigating Adversarial Attacks in Software Analytics via Machine Learning Explainability. (99%)
Enhancing Output Diversity Improves Conjugate Gradient-based Adversarial Attacks. (98%)
EdgeShield: A Universal and Efficient Edge Computing Framework for Robust AI. (83%)
EnJa: Ensemble Jailbreak on Large Language Models. (83%)
MORTAR: A Model-based Runtime Action Repair Framework for AI-enabled Cyber-Physical Systems. (76%)
LaFA: Latent Feature Attacks on Non-negative Matrix Factorization. (38%)
MTDSense: AI-Based Fingerprinting of Moving Target Defense Techniques in Software-Defined Networking. (26%)
FDI: Attack Neural Code Generation Systems through User Feedback Channel. (5%)
Hard to Explain: On the Computational Hardness of In-Distribution Model Interpretation. (1%)
Decoding Biases: Automated Methods and LLM Judges for Gender Bias Detection in Language Models. (1%)
2024-08-06
Adversarial Robustness of Open-source Text Classification Models and Fine-Tuning Chains. (98%)
Sample-agnostic Adversarial Perturbation for Vision-Language Pre-training Models. (98%)
Simple Perturbations Subvert Ethereum Phishing Transactions Detection: An Empirical Analysis. (92%)
Attacks and Defenses for Generative Diffusion Models: A Comprehensive Survey. (64%)
A Study on Prompt Injection Attack Against LLM-Integrated Mobile Robotic Systems. (2%)
2024-08-05
On the Robustness of Malware Detectors to Adversarial Samples. (99%)
Mitigating Malicious Attacks in Federated Learning via Confidence-aware Defense. (84%)
SEAS: Self-Evolving Adversarial Safety Optimization for Large Language Models. (26%)
Why Are My Prompts Leaked? Unraveling Prompt Extraction Threats in Customized Large Language Models. (13%)
Pre-trained Encoder Inference: Revealing Upstream Encoders In Downstream Machine Learning Services. (13%)
Can Reinforcement Learning Unlock the Hidden Dangers in Aligned Large Language Models? (8%)
RCDM: Enabling Robustness for Conditional Diffusion Model. (4%)
Compromising Embodied Agents with Contextual Backdoor Attacks. (4%)
Practical Attacks against Black-box Code Completion Engines. (4%)
2024-08-04
A Survey and Evaluation of Adversarial Attacks for Object Detection. (99%)
AdvQDet: Detecting Query-Based Adversarial Attacks with Adversarial Contrastive Prompt Tuning. (99%)
Label Augmentation for Neural Networks Robustness. (98%)
Top K Enhanced Reinforcement Learning Attacks on Heterogeneous Graph Node Classification. (76%)
Model Hijacking Attack in Federated Learning. (75%)
Robustness of Watermarking on Text-to-Image Diffusion Models. (22%)
FovEx: Human-inspired Explanations for Vision Transformers and Convolutional Neural Networks. (1%)
2024-08-03
ALIF: Low-Cost Adversarial Audio Attacks on Black-Box Speech Platforms using Linguistic Features. (99%)
Joint Universal Adversarial Perturbations with Interpretations. (99%)
Downstream Transfer Attack: Adversarial Attacks on Downstream Models with Pre-trained Vision Transformers. (99%)
2024-08-02
Guardians of Image Quality: Benchmarking Defenses Against Adversarial Attacks on Image Quality Metrics. (98%)
Trustworthy Machine Learning under Social and Adversarial Data Sources. (83%)
EmoBack: Backdoor Attacks Against Speaker Identification Using Emotional Prosody. (80%)
Interpreting Global Perturbation Robustness of Image Models using Axiomatic Spectral Importance Decomposition. (61%)
Assessing Robustness of Machine Learning Models using Covariate Perturbations. (33%)
Certifiably Robust Encoding Schemes. (31%)
Hallu-PI: Evaluating Hallucination in Multi-modal Large Language Models within Perturbed Inputs. (2%)
2024-08-01
Autonomous LLM-Enhanced Adversarial Attack for Text-to-Motion. (99%)
OTAD: An Optimal Transport-Induced Robust Model for Agnostic Adversarial Attack. (99%)
Securing the Diagnosis of Medical Imaging: An In-depth Analysis of AI-Resistant Attacks. (99%)
CERT-ED: Certifiably Robust Text Classification for Edit Distance. (98%)
ADBM: Adversarial diffusion bridge model for reliable adversarial purification. (96%)
Discrete Randomized Smoothing Meets Quantum Computing. (41%)
Adversarial Text Rewriting for Text-aware Recommender Systems. (13%)
MAARS: Multi-Rate Attack-Aware Randomized Scheduling for Securing Real-time Systems. (1%)
Pathway to Secure and Trustworthy 6G for LLMs: Attacks, Defense, and Opportunities. (1%)
2024-07-31
Cross-modality Information Check for Detecting Jailbreaking in Multimodal Large Language Models. (98%)
On the Perturbed States for Transformed Input-robust Reinforcement Learning. (92%)
The Llama 3 Herd of Models. (62%)
Certifying Robustness of Learning-Based Keypoint Detection and Pose Estimation Methods. (22%)
Vera Verto: Multimodal Hijacking Attack. (9%)
2024-07-30
Prompt-Driven Contrastive Learning for Transferable Adversarial Attacks. (99%)
AI Safety in Practice: Enhancing Adversarial Robustness in Multimodal Image Captioning. (99%)
FACL-Attack: Frequency-Aware Contrastive Learning for Transferable Adversarial Attacks. (99%)
Vulnerabilities in AI-generated Image Detection: The Challenge of Adversarial Attacks. (99%)
Diff-Cleanse: Identifying and Mitigating Backdoor Attacks in Diffusion Models. (62%)
DeepBaR: Fault Backdoor Attack on Deep Neural Network Layers. (47%)
Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification. (16%)
Bayesian Low-Rank LeArning (Bella): A Practical Approach to Bayesian Neural Networks. (1%)
2024-07-29
Adversarial Robustness in RGB-Skeleton Action Recognition: Leveraging Attention Modality Reweighter. (99%)
Enhancing Adversarial Text Attacks on BERT Models with Projected Gradient Descent. (99%)
Detecting and Understanding Vulnerabilities in Language Models via Mechanistic Interpretability. (92%)
From ML to LLM: Evaluating the Robustness of Phishing Webpage Detection Models against Adversarial Attacks. (84%)
DDAP: Dual-Domain Anti-Personalization against Text-to-Image Diffusion Models. (68%)
RSC-SNN: Exploring the Trade-off Between Adversarial Robustness and Accuracy in Spiking Neural Networks via Randomized Smoothing Coding. (50%)
Can Editing LLMs Inject Harm? (9%)
BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learning. (3%)
ImagiNet: A Multi-Content Dataset for Generalizable Synthetic Image Detection via Contrastive Learning. (1%)
2024-07-28
Exploring the Adversarial Robustness of CLIP for AI-generated Image Detection. (80%)
2024-07-27
EaTVul: ChatGPT-based Evasion Attack Against Software Vulnerability Detection. (99%)
Towards Clean-Label Backdoor Attacks in the Physical World. (98%)
2024-07-26
Debiased Graph Poisoning Attack via Contrastive Surrogate Objective. (93%)
Robust VAEs via Generating Process of Noise Augmented Data. (87%)
Adversarial Robustification via Text-to-Image Diffusion Models. (64%)
A Survey of Malware Detection Using Deep Learning. (5%)
Unveiling Privacy Vulnerabilities: Investigating the Role of Structure in Graph Data. (1%)
UniForensics: Face Forgery Detection via General Facial Representation. (1%)
2024-07-25
Sparse vs Contiguous Adversarial Pixel Perturbations in Multimodal Models: An Empirical Analysis. (99%)
Effects of Scale on Language Model Robustness. (96%)
A Unified Understanding of Adversarial Vulnerability Regarding Unimodal Models and Vision-Language Pre-training Models. (95%)
RIDA: A Robust Attack Framework on Incomplete Graphs. (31%)
Adversarially Robust Decision Transformer. (22%)
Peak-Controlled Logits Poisoning Attack in Federated Distillation. (4%)
Network Inversion of Convolutional Neural Nets. (3%)
Regret-Optimal Defense Against Stealthy Adversaries: A System Level Approach. (1%)
2024-07-24
Physical Adversarial Attack on Monocular Depth Estimation via Shape-Varying Patches. (92%)
FLRT: Fluent Student-Teacher Redteaming. (13%)
2024-07-23
S-E Pipeline: A Vision Transformer (ViT) based Resilient Classification Pipeline for Medical Imaging Against Adversarial Attacks. (87%)
Algebraic Adversarial Attacks on Integrated Gradients. (86%)
Multimodal Unlearnable Examples: Protecting Data against Multimodal Contrastive Learning. (41%)
When AI Defeats Password Deception! A Deep Learning Framework to Distinguish Passwords and Honeywords. (13%)
Figure it Out: Analyzing-based Jailbreak Attack on Large Language Models. (8%)
RedAgent: Red Teaming Large Language Models with Context-aware Autonomous Language Agent. (5%)
2024-07-22
Enhancing Transferability of Targeted Adversarial Examples: A Self-Universal Perspective. (99%)
Towards Robust Vision Transformer via Masked Adaptive Ensemble. (99%)
Towards Efficient Transferable Preemptive Adversarial Defense. (99%)
On Feasibility of Intent Obfuscating Attacks. (98%)
Poisoning with A Pill: Circumventing Detection in Federated Learning. (92%)
Revisiting the Robust Alignment of Circuit Breakers. (70%)
Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs. (56%)
Imposter.AI: Adversarial Attacks with Hidden Intentions towards Aligned Large Language Models. (11%)
Virtual Reality and Augmented Reality Security: A Reconnaissance and Vulnerability Assessment Approach. (1%)
2024-07-21
Taxonomy Driven Fast Adversarial Training. (99%)
When Do Universal Image Jailbreaks Transfer Between Vision-Language Models? (74%)
A Learning-Based Attack Framework to Break SOTA Poisoning Defenses in Federated Learning. (73%)
SeqMIA: Sequential-Metric Based Membership Inference Attack. (22%)
Explainable AI-based Intrusion Detection System for Industry 5.0: An Overview of the Literature, associated Challenges, the existing Solutions, and Potential Research Directions. (5%)
Assessing Brittleness of Image-Text Retrieval Benchmarks from Vision-Language Models Perspective. (2%)
2024-07-20
Sim-CLIP: Unsupervised Siamese Adversarial Fine-Tuning for Robust and Semantically-Rich Vision-Language Models. (68%)
2024-07-19
Data Poisoning: An Overlooked Threat to Power Grid Resilience. (68%)
Human-Interpretable Adversarial Prompt Attack on Large Language Models with Situational Context. (4%)
Adversarial Databases Improve Success in Retrieval-based Large Language Models. (1%)
On the Robustness of Fully-Spiking Neural Networks in Open-World Scenarios using Forward-Only Learning Algorithms. (1%)
2024-07-18
Cross-Task Attack: A Self-Supervision Generative Framework Based on Attention Shift. (99%)
Beyond Dropout: Robust Convolutional Neural Networks Based on Local Feature Masking. (98%)
Black-Box Opinion Manipulation Attacks to Retrieval-Augmented Generation of Large Language Models. (75%)
Prover-Verifier Games improve legibility of LLM outputs. (61%)
Compressed models are NOT miniature versions of large models. (47%)
Distributionally and Adversarially Robust Logistic Regression via Intersecting Wasserstein Balls. (16%)
A Closer Look at GAN Priors: Exploiting Intermediate Features for Enhanced Model Inversion Attacks. (10%)
2024-07-17
PG-Attack: A Precision-Guided Adversarial Attack Framework Against Vision Foundation Models for Autonomous Driving. (98%)
Preventing Catastrophic Overfitting in Fast Adversarial Training: A Bi-level Optimization Perspective. (98%)
Transferable Adversarial Facial Images for Privacy Protection. (96%)
Context-Aware Fuzzing for Robustness Enhancement of Deep Learning Models. (86%)
Krait: A Backdoor Attack Against Graph Prompt Tuning. (83%)
AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases. (61%)
Benchmarking Robust Self-Supervised Learning Across Diverse Downstream Tasks. (12%)
Direct Unlearning Optimization for Robust and Safe Text-to-Image Models. (12%)
Contrastive Adversarial Training for Unsupervised Domain Adaptation. (2%)
Rethinking Video-Text Understanding: Retrieval from Counterfactually Augmented Data. (1%)
2024-07-16
Any Target Can be Offense: Adversarial Example Generation via Generalized Latent Infection. (99%)
Variational Randomized Smoothing for Sample-Wise Adversarial Robustness. (99%)
AEMIM: Adversarial Examples Meet Masked Image Modeling. (99%)
Investigating Imperceptibility of Adversarial Attacks on Tabular Data: An Empirical Analysis. (99%)
Enhancing TinyML Security: Study of Adversarial Attack Transferability. (96%)
UNIT: Backdoor Mitigation via Automated Neural Distribution Tightening. (82%)
Relaxing Graph Transformers for Adversarial Attacks. (81%)
Turning Generative Models Degenerate: The Power of Data Poisoning Attacks. (76%)
Learning on Graphs with Large Language Models(LLMs): A Deep Dive into Model Robustness. (33%)
SegSTRONG-C: Segmenting Surgical Tools Robustly On Non-adversarial Generated Corruptions -- An EndoVis'24 Challenge. (33%)
Does Refusal Training in LLMs Generalize to the Past Tense? (15%)
Cover-separable Fixed Neural Network Steganography via Deep Generative Models. (8%)
Model Inversion Attacks Through Target-Specific Conditional Diffusion Models. (4%)
IPA-NeRF: Illusory Poisoning Attack Against Neural Radiance Fields. (1%)
2024-07-15
Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks. (99%)
Backdoor Attacks against Image-to-Image Networks. (88%)
Towards Adversarially Robust Vision-Language Models: Insights from Design Choices and Prompt Formatting Techniques. (88%)
PartImageNet++ Dataset: Scaling up Part-based Models for Robust Recognition. (80%)
Provable Robustness of (Graph) Neural Networks Against Data Poisoning and Backdoor Attacks. (67%)
Uncertainty is Fragile: Manipulating Uncertainty in Large Language Models. (41%)
Feature Inference Attack on Shapley Values. (12%)
2024-07-14
Transferable 3D Adversarial Shape Completion using Diffusion Models. (99%)
Towards Robust Recommendation via Decision Boundary-aware Graph Contrastive Learning. (92%)
Defending Against Repetitive-based Backdoor Attacks on Semi-supervised Learning through Lens of Rate-Distortion-Perception Trade-off. (76%)
CLIP-Guided Networks for Transferable Targeted Attacks. (76%)
SENTINEL: Securing Indoor Localization against Adversarial Attacks with Capsule Neural Networks. (10%)
2024-07-13
Augmented Neural Fine-Tuning for Efficient Backdoor Purification. (68%)
Partner in Crime: Boosting Targeted Poisoning Attacks against Federated Learning. (67%)
Team up GBDTs and DNNs: Advancing Efficient and Effective Tabular Prediction with Tree-hybrid MLPs. (1%)
2024-07-12
SemiAdv: Query-Efficient Black-Box Adversarial Attack with Unlabeled Images. (99%)
Evaluating the Adversarial Robustness of Semantic Segmentation: Trying Harder Pays Off. (97%)
TAPI: Towards Target-Specific and Adversarial Prompt Injection against Code LLMs. (93%)
Deep Adversarial Defense Against Multilevel-Lp Attacks. (87%)
Robust Yet Efficient Conformal Prediction Sets. (61%)
Refusing Safe Prompts for Multi-modal Large Language Models. (16%)
Security Matrix for Multimodal Agents on Mobile Devices: A Systematic and Proof of Concept Study. (15%)
MaPPing Your Model: Assessing the Impact of Adversarial Attacks on LLM-based Programming Assistants. (13%)
BoBa: Boosting Backdoor Detection through Data Distribution Inference in Federated Learning. (5%)
Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training. (1%)
2024-07-11
Rethinking the Threat and Accessibility of Adversarial Attacks against Face Recognition Systems. (99%)
Boosting Adversarial Transferability for Skeleton-based Action Recognition via Exploring the Model Posterior Space. (99%)
HO-FMN: Hyperparameter Optimization for Fast Minimum-Norm Attacks. (98%)
Distributed Backdoor Attacks on Federated Graph Learning and Certified Defenses. (98%)
DeCE: Deceptive Cross-Entropy Loss Designed for Defending Backdoor Attacks. (87%)
How to beat a Bayesian adversary. (81%)
Soft Prompts Go Hard: Steering Visual Language Models with Hidden Meta-Instructions. (74%)
DART: A Solution for Decentralized Federated Learning Model Robustness Analysis. (47%)
Quantitative Evaluation of the Saliency Map for Alzheimer's Disease Classifier with Anatomical Segmentation. (8%)
Enhancing Privacy of Spatiotemporal Federated Learning against Gradient Inversion Attacks. (8%)
Are Large Language Models Really Bias-Free? Jailbreak Prompts for Assessing Adversarial Robustness to Bias Elicitation. (1%)
Deep Learning for Network Anomaly Detection under Data Contamination: Evaluating Robustness and Mitigating Performance Degradation. (1%)
2024-07-10
Adversarial Attacks and Defenses on Text-to-Image Diffusion Models: A Survey. (99%)
A Survey of Attacks on Large Vision-Language Models: Resources, Advances, and Future Trends. (38%)
Model-agnostic clean-label backdoor mitigation in cybersecurity environments. (31%)
Flooding Spread of Manipulated Knowledge in LLM-Based Multi-Agent Communities. (11%)
Invisible Optical Adversarial Stripes on Traffic Sign against Autonomous Vehicles. (8%)
A Comprehensive Survey on the Security of Smart Grid: Challenges, Mitigations, and Future Research Opportunities. (2%)
Was it Slander? Towards Exact Inversion of Generative Language Models. (2%)
CHILLI: A data context-aware perturbation method for XAI. (1%)
2024-07-09
A Hybrid Training-time and Run-time Defense Against Adversarial Attacks in Modulation Classification. (99%)
Universal Multi-view Black-box Attack against Object Detectors via Layout Optimization. (99%)
DLOVE: A new Security Evaluation Tool for Deep Learning Based Watermarking Techniques. (98%)
Countermeasures Against Adversarial Examples in Radio Signal Classification. (97%)
Improving the Transferability of Adversarial Examples by Feature Augmentation. (93%)
Tracing Back the Malicious Clients in Poisoning Attacks to Federated Learning. (26%)
The Quantum Imitation Game: Reverse Engineering of Quantum Machine Learning Models. (15%)
Robust Neural Information Retrieval: An Adversarial and Out-of-distribution Perspective. (13%)
Attack GAN (AGAN ): A new Security Evaluation Tool for Perceptual Encryption. (10%)
Performance Evaluation of Knowledge Graph Embedding Approaches under Non-adversarial Attacks. (8%)
Exploring the Causality of End-to-End Autonomous Driving. (1%)
Distribution System Reconfiguration to Mitigate Load Altering Attacks via Stackelberg Games. (1%)
2024-07-08
Shedding More Light on Robust Classifiers under the lens of Energy-based Models. (98%)
Non-Robust Features are Not Always Useful in One-Class Classification. (92%)
Exposing Privacy Gaps: Membership Inference Attack on Preference Data for LLM Alignment. (1%)
2024-07-07
Rethinking Targeted Adversarial Attacks For Neural Machine Translation. (99%)
Gradient Diffusion: A Perturbation-Resilient Gradient Leakage Attack. (64%)
An accurate detection is not all you need to combat label noise in web-noisy datasets. (1%)
Detecting new obfuscated malware variants: A lightweight and interpretable machine learning approach. (1%)
Evolutionary Trigger Detection and Lightweight Model Repair Based Backdoor Defense. (1%)
2024-07-06
A Novel Bifurcation Method for Observation Perturbation Attacks on Reinforcement Learning Agents: Load Altering Attacks on a Cyber Physical Power System. (99%)
Releasing Malevolence from Benevolence: The Menace of Benign Data on Machine Unlearning. (13%)
GCON: Differentially Private Graph Convolutional Network via Objective Perturbation. (12%)
2024-07-05
Remembering Everything Makes You Vulnerable: A Limelight on Machine Unlearning for Personalized Healthcare Sector. (98%)
Jailbreak Attacks and Defenses Against Large Language Models: A Survey. (92%)
Controlling Whisper: Universal Acoustic Adversarial Attacks to Control Speech Foundation Models. (91%)
Self-Supervised Representation Learning for Adversarial Attack Detection. (68%)
On Evaluating The Performance of Watermarked Machine-Generated Texts Under Adversarial Attacks. (61%)
Late Breaking Results: Fortifying Neural Networks: Safeguarding Against Adversarial Attacks with Stochastic Computing. (54%)
Regulating Model Reliance on Non-Robust Features by Smoothing Input Marginal Density. (38%)
Data Poisoning Attacks in Intelligent Transportation Systems: A Survey. (2%)
Non-Cooperative Backdoor Attacks in Federated Learning: A New Threat Landscape. (2%)
2024-07-04
TrackPGD: A White-box Attack using Binary Masks against Robust Transformer Trackers. (99%)
Protecting Deep Learning Model Copyrights with Adversarial Example-Free Reuse Detection. (99%)
Adversarial Robustness of VAEs across Intersectional Subgroups. (99%)
Mitigating Low-Frequency Bias: Feature Recalibration and Frequency Attention Regularization for Adversarial Robustness. (92%)
Securing Multi-turn Conversational Language Models Against Distributed Backdoor Triggers. (68%)
Charging Ahead: A Hierarchical Adversarial Framework for Counteracting Advanced Cyber Threats in EV Charging Stations. (15%)
T2IShield: Defending Against Backdoors on Text-to-Image Diffusion Models. (13%)
Automated Progressive Red Teaming. (2%)
Quantifying Prediction Consistency Under Model Multiplicity in Tabular LLMs. (1%)
Certifiably Robust Image Watermark. (1%)
2024-07-03
A Wolf in Sheep's Clothing: Practical Black-box Adversarial Attacks for Evading Learning-based Windows Malware Detection in the Wild. (99%)
$L_p$-norm Distortion-Efficient Adversarial Attack. (99%)
SPLITZ: Certifiable Robustness via Split Lipschitz Randomized Smoothing. (98%)
JailbreakHunter: A Visual Analytics Approach for Jailbreak Prompts Discovery from Large-Scale Human-LLM Conversational Datasets. (83%)
Venomancer: Towards Imperceptible and Target-on-Demand Backdoor Attacks in Federated Learning. (74%)
A Geometric Framework for Adversarial Vulnerability in Machine Learning. (70%)
Self-Evaluation as a Defense Against Adversarial Attacks on LLMs. (41%)
Backdoor Graph Condensation. (16%)
Safe Unlearning: A Surprisingly Effective and Generalizable Solution to Defend Against Jailbreak Attacks. (10%)
Federated Learning for Zero-Day Attack Detection in 5G and Beyond V2X Networks. (2%)
An Empirical Study on Capability of Large Language Models in Understanding Code Semantics. (1%)
On Large Language Models in National Security Applications. (1%)
Purification Of Contaminated Convolutional Neural Networks Via Robust Recovery: An Approach with Theoretical Guarantee in One-Hidden-Layer Case. (1%)
2024-07-02
Secure Semantic Communication via Paired Adversarial Residual Networks. (99%)
EvolBA: Evolutionary Boundary Attack under Hard-label Black Box condition. (99%)
Adversarial Magnification to Deceive Deepfake Detection through Super Resolution. (98%)
Breach By A Thousand Leaks: Unsafe Information Leakage in `Safe' AI Responses. (80%)
Light-weight Fine-tuning Method for Defending Adversarial Noise in Pre-trained Medical Vision-Language Models. (76%)
Parameter Matching Attack: Enhancing Practical Applicability of Availability Attacks. (50%)
Towards More Realistic Extraction Attacks: An Adversarial Perspective. (22%)
On the Robustness of Graph Reduction Against GNN Backdoor. (13%)
MALT Powers Up Adversarial Attacks. (13%)
Face Reconstruction Transfer Attack as Out-of-Distribution Generalization. (2%)
Robust ADAS: Enhancing Robustness of Machine Learning-based Advanced Driver Assistance Systems for Adverse Weather. (1%)
2024-07-01
Multi-View Black-Box Physical Attacks on Infrared Pedestrian Detectors Using Adversarial Infrared Grid. (98%)
DeepiSign-G: Generic Watermark to Stamp Hidden DNN Parameters for Self-contained Tracking. (82%)
Looking From the Future: Multi-order Iterations Can Enhance Adversarial Attack Transferability. (81%)
QUEEN: Query Unlearning against Model Extraction. (75%)
Formal Verification of Object Detection. (56%)
SoP: Unlock the Power of Social Facilitation for Automatic Jailbreak Attack. (13%)
Securing Distributed Network Digital Twin Systems Against Model Poisoning Attacks. (8%)
A Fingerprint for Large Language Models. (2%)
Unveiling the Unseen: Exploring Whitebox Membership Inference through the Lens of Explainability. (1%)
Unaligning Everything: Or Aligning Any Text to Any Image in Multimodal Models. (1%)
2024-06-30
Learning Robust 3D Representation from CLIP via Dual Denoising. (67%)
Consistency Purification: Effective and Efficient Diffusion Purification towards Certified Robustness. (13%)
UWBAD: Towards Effective and Imperceptible Jamming Attacks Against UWB Ranging Systems with COTS Chips. (2%)
2024-06-29
Query-Efficient Hard-Label Black-Box Attack against Vision Transformers. (99%)
2024-06-28
Deceptive Diffusion: Generating Synthetic Adversarial Examples. (99%)
DiffuseDef: Improved Robustness to Adversarial Attacks. (95%)
Emotion Loss Attacking: Adversarial Attack Perception for Skeleton based on Multi-dimensional Features. (92%)
Steering cooperation: Adversarial attacks on prisoner's dilemma in complex networks. (92%)
IDT: Dual-Task Adversarial Attacks for Privacy Protection. (88%)
Backdoor Attack in Prompt-Based Continual Learning. (16%)
Virtual Context: Enhancing Jailbreak Attacks with Special Token Injection. (11%)
GRACE: Graph-Regularized Attentive Convolutional Entanglement with Laplacian Smoothing for Robust DeepFake Video Detection. (1%)
2024-06-27
Zero-Query Adversarial Attack on Black-box Automatic Speech Recognition Systems. (99%)
Data-Driven Lipschitz Continuity: A Cost-Effective Approach to Improve Adversarial Robustness. (98%)
Investigating and Defending Shortcut Learning in Personalized Diffusion Models. (87%)
Data Poisoning Attacks to Locally Differentially Private Frequent Itemset Mining Protocols. (2%)
Context Matters: An Empirical Study of the Impact of Contextual Information in Temporal Question Answering Systems. (1%)
2024-06-26
Detecting Brittle Decisions for Free: Leveraging Margin Consistency in Deep Robust Classifiers. (98%)
On Discrete Prompt Optimization for Diffusion Models. (62%)
Breaking the Barrier: Enhanced Utility and Robustness in Smoothed DRL Agents. (54%)
Poisoned LangChain: Jailbreak LLMs by LangChain. (26%)
Revisiting Backdoor Attacks against Large Vision-Language Models. (15%)
WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models. (12%)
Adversarial Search Engine Optimization for Large Language Models. (9%)
2024-06-25
CuDA2: An approach for Incorporating Traitor Agents into Cooperative Multi-Agent Systems. (99%)
Diffusion-based Adversarial Purification for Intrusion Detection. (98%)
Semantic Deep Hiding for Robust Unlearnable Examples. (76%)
Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness. (67%)
Robustly Optimized Deep Feature Decoupling Network for Fatty Liver Diseases Detection. (13%)
2024-06-24
Evaluating the Robustness of Deep-Learning Algorithm-Selection Models by Evolving Adversarial Instances. (98%)
UNICAD: A Unified Approach for Attack Detection, Noise Reduction and Novel Class Identification. (96%)
ADVSCORE: A Metric for the Evaluation and Creation of Adversarial Benchmarks. (92%)
Automated Adversarial Discovery for Safety Classifiers. (92%)
Improving robustness to corruptions with multiplicative weight perturbations. (74%)
BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models. (38%)
From Perfect to Noisy World Simulation: Customizable Embodied Multi-modal Perturbations for SLAM Robustness Benchmarking. (5%)
Machine Unlearning Fails to Remove Data Poisoning Attacks. (2%)
2024-06-23
Towards unlocking the mystery of adversarial fragility of neural networks. (64%)
CBPF: Filtering Poisoned Data Based on Composite Backdoor Attack. (13%)
Investigating the Influence of Prompt-Specific Shortcuts in AI Generated Text Detection. (8%)
On Instabilities of Unsupervised Denoising Diffusion Models in Magnetic Resonance Imaging Reconstruction. (2%)
Understanding and Diagnosing Deep Reinforcement Learning. (1%)
2024-06-22
The Effect of Similarity Measures on Accurate Stability Estimates for Local Surrogate Models in Text-based Explainable AI. (97%)
Federated Adversarial Learning for Robust Autonomous Landing Runway Detection. (2%)
Privacy Implications of Explainable AI in Data-Driven Systems. (1%)
2024-06-21
ECLIPSE: Expunging Clean-label Indiscriminate Poisons via Sparse Diffusion Purification. (99%)
Deciphering the Definition of Adversarial Robustness for post-hoc OOD Detectors. (99%)
DataFreeShield: Defending Adversarial Attacks without Training Data. (45%)
Large Language Models for Link Stealing Attacks Against Graph Neural Networks. (38%)
Logicbreaks: A Framework for Understanding Subversion of Rule-based Inference. (2%)
MOUNTAINEER: Topology-Driven Visual Analytics for Comparing Local Explanations. (1%)
2024-06-20
Enhancing robustness of data-driven SHM models: adversarial training with circle loss. (99%)
Exploring Layerwise Adversarial Robustness Through the Lens of t-SNE. (87%)
Defending Against Sophisticated Poisoning Attacks with RL-based Aggregation in Federated Learning. (81%)
Jailbreaking as a Reward Misspecification Problem. (78%)
Uniform Convergence of Adversarially Robust Classifiers. (68%)
Prompt Injection Attacks in Defended Systems. (47%)
MEAT: Median-Ensemble Adversarial Training for Improving Robustness and Generalization. (41%)
Countering adversarial perturbations in graphs using error correcting codes. (22%)
Steering Without Side Effects: Improving Post-Deployment Control of Language Models. (15%)
Evaluating Implicit Bias in Large Language Models by Attacking From a Psychometric Perspective. (8%)
PoseBench: Benchmarking the Robustness of Pose Estimation Models under Corruptions. (5%)
Can you trust your explanations? A robustness test for feature attribution methods. (2%)
SeCTIS: A Framework to Secure CTI Sharing. (1%)
2024-06-19
GraphMU: Repairing Robustness of Graph Neural Networks via Machine Unlearning. (99%)
AGSOA:Graph Neural Network Targeted Attack Based on Average Gradient and Structure Optimization. (99%)
Explainable AI Security: Exploring Robustness of Graph Neural Networks to Adversarial Attacks. (99%)
AgentDojo: A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents. (83%)
Enhancing Cross-Prompt Transferability in Vision-Language Models through Contextual Injection of Target Tokens. (62%)
Textual Unlearning Gives a False Sense of Unlearning. (16%)
Large-Scale Dataset Pruning in Adversarial Training through Data Importance Extrapolation. (9%)
DPO: Dual-Perturbation Optimization for Test-time Adaptation in 3D Object Detection. (3%)
ModSec-Learn: Boosting ModSecurity with Machine Learning. (2%)
RobGC: Towards Robust Graph Condensation. (1%)
2024-06-18
Saliency Attention and Semantic Similarity-Driven Adversarial Perturbation. (99%)
NoiSec: Harnessing Noise for Security against Adversarial and Backdoor Attacks. (97%)
MaskPure: Improving Defense Against Text Adversaries with Stochastic Purification. (95%)
Adversarial Attacks on Multimodal Agents. (91%)
Towards Trustworthy Unsupervised Domain Adaptation: A Representation Learning Perspective for Enhancing Robustness, Discrimination, and Generalization. (76%)
Adversarial Attacks on Large Language Models in Medicine. (70%)
Can Go AIs be adversarially robust? (61%)
DLP: towards active defense against backdoor attacks with decoupled learning process. (31%)
Attack and Defense of Deep Learning Models in the Field of Web Attack Detection. (10%)
SHIELD: Evaluation and Defense Strategies for Copyright Compliance in LLM Text Generation. (10%)
CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models. (8%)
Stealth edits for provably fixing or attacking large language models. (2%)
PRePair: Pointwise Reasoning Enhance Pairwise Evaluating for Robust Instruction-Following Assessments. (1%)
2024-06-17
Obfuscating IoT Device Scanning Activity via Adversarial Example Generation. (99%)
FullCert: Deterministic End-to-End Certification for Training and Inference of Neural Networks. (98%)
Harmonizing Feature Maps: A Graph Convolutional Approach for Enhancing Adversarial Robustness. (93%)
A First Physical-World Trajectory Prediction Attack via LiDAR-induced Deceptions in Autonomous Driving. (82%)
Adversarial Perturbations Cannot Reliably Protect Artists From Generative AI. (76%)
ToxiCloakCN: Evaluating Robustness of Offensive Language Detection in Chinese with Cloaking Perturbations. (22%)
Evading AI-Generated Content Detectors using Homoglyphs. (3%)
BadSampler: Harnessing the Power of Catastrophic Forgetting to Poison Byzantine-robust Federated Learning. (2%)
SoK: A Literature and Engineering Review of Regular Expression Denial of Service. (2%)
Do Parameters Reveal More than Loss for Membership Inference? (1%)
Adversaries With Incentives: A Strategic Alternative to Adversarial Robustness. (1%)
2024-06-16
Improving Adversarial Robustness via Decoupled Visual Representation Masking. (99%)
Imperceptible Face Forgery Attack via Adversarial Semantic Mask. (99%)
KGPA: Robustness Evaluation for Large Language Models via Cross-Domain Knowledge Graphs. (92%)
NBA: defensive distillation for backdoor removal via neural behavior alignment. (80%)
RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models. (62%)
ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates. (61%)
Imperceptible Rhythm Backdoor Attacks: Exploring Rhythm Transformation for Embedding Undetectable Vulnerabilities on Speech Recognition. (10%)
RUPBench: Benchmarking Reasoning Under Perturbations for Robustness Evaluation in Large Language Models. (9%)
2024-06-15
Robust Image Classification in the Presence of Out-of-Distribution and Adversarial Samples Using Attractors in Neural Networks. (98%)
E-SAGE: Explainability-based Defense Against Backdoor Attacks on Graph Neural Networks. (81%)
Emerging Safety Attack and Defense in Federated Instruction Tuning of Large Language Models. (68%)
Enhancing Anomaly Detection Generalization through Knowledge Exposure: The Dual Effects of Augmentation. (1%)
2024-06-14
Over-parameterization and Adversarial Robustness in Neural Networks: An Overview and Empirical Analysis. (99%)
Adaptive Randomized Smoothing: Certified Adversarial Robustness for Multi-Step Defences. (93%)
Robustness-Inspired Defense Against Backdoor Attacks on Graph Neural Networks. (75%)
Automated Design of Linear Bounding Functions for Sigmoidal Nonlinearities in Neural Networks. (67%)
Beyond Slow Signs in High-fidelity Model Extraction. (10%)
Byzantine-Robust Decentralized Federated Learning. (8%)
2024-06-13
Improving Adversarial Robustness via Feature Pattern Consistency Constraint. (99%)
Watch the Watcher! Backdoor Attacks on Security-Enhancing Diffusion Models. (98%)
MirrorCheck: Efficient Adversarial Defense for Vision-Language Models. (95%)
Towards Evaluating the Robustness of Visual State Space Models. (89%)
Bag of Tricks: Benchmarking of Jailbreak Attacks on LLMs. (11%)
Steganalysis on Digital Watermarking: Is Your Defense Truly Impervious? (4%)
Validation of human benchmark models for Automated Driving System approval: How competent and careful are they really? (1%)
An Unsupervised Approach to Achieve Supervised-Level Explainability in Healthcare Records. (1%)
Large-Scale Evaluation of Open-Set Image Classification Techniques. (1%)
Understanding Hallucinations in Diffusion Models through Mode Interpolation. (1%)
2024-06-12
I Don't Know You, But I Can Catch You: Real-Time Defense against Diverse Adversarial Patches for Object Detectors. (99%)
On Evaluating Adversarial Robustness of Volumetric Medical Segmentation Models. (99%)
Adversarial Evasion Attack Efficiency against Large Language Models. (98%)
Transformation-Dependent Adversarial Attacks. (89%)
When LLM Meets DRL: Advancing Jailbreaking Efficiency via DRL-guided Search. (64%)
RL-JACK: Reinforcement Learning-powered Black-box Jailbreaking Attack against LLMs. (62%)
AdaNCA: Neural Cellular Automata As Adaptors For More Robust Vision Transformer. (22%)
Graph Transductive Defense: a Two-Stage Defense for Graph Membership Inference Attacks. (13%)
On Security Weaknesses and Vulnerabilities in Deep Learning Systems. (8%)
Dataset and Lessons Learned from the 2024 SaTML LLM Capture-the-Flag Competition. (4%)
Improving Noise Robustness through Abstractions and its Impact on Machine Learning. (4%)
Exploiting Uncommon Text-Encoded Structures for Automated Jailbreaks in LLMs. (1%)
Adversarial Patch for 3D Local Feature Extractor. (1%)
2024-06-11
Erasing Radio Frequency Fingerprints via Active Adversarial Perturbation. (86%)
AudioMarkBench: Benchmarking Robustness of Audio Watermarking. (83%)
On the H\"{o}lder Stability of Multiset and Graph Neural Networks. (69%)
A Study of Backdoors in Instruction Fine-tuned Language Models. (31%)
Merging Improves Self-Critique Against Jailbreak Attacks. (26%)
Benchmarking Trustworthiness of Multimodal Large Language Models: A Comprehensive Study. (15%)
Dual Thinking and Perceptual Analysis of Deep Learning Models using Human Adversarial Examples. (15%)
MoreauPruner: Robust Pruning of Large Language Models against Weight Perturbations. (5%)
Rethinking the impact of noisy labels in graph classification: A utility and privacy perspective. (1%)
Agnostic Sharpness-Aware Minimization. (1%)
2024-06-10
Texture Re-scalable Universal Adversarial Perturbation. (99%)
Explainable Graph Neural Networks Under Fire. (99%)
Lurking in the shadows: Unveiling Stealthy Backdoor Attacks against Personalized Federated Learning. (81%)
Reinforced Compressive Neural Architecture Search for Versatile Adversarial Robustness. (56%)
Raccoon: Prompt Extraction Benchmark of LLM-Integrated Applications. (56%)
A Survey of Backdoor Attacks and Defenses on Large Language Models: Implications for Security Measures. (13%)
An LLM-Assisted Easy-to-Trigger Backdoor Attack on Code Completion Models: Injecting Disguised Vulnerabilities against Strong Detection. (8%)
Fast White-Box Adversarial Streaming Without a Random Oracle. (3%)
Unveiling the Safety of GPT-4o: An Empirical Study using Jailbreak Attacks. (2%)
2024-06-09
DMS: Addressing Information Loss with More Steps for Pragmatic Adversarial Attacks. (99%)
MeanSparse: Post-Training Robustness Enhancement Through Mean-Centered Feature Sparsification. (97%)
Stealthy Targeted Backdoor Attacks against Image Captioning. (82%)
ControlLoc: Physical-World Hijacking Attack on Visual Perception in Autonomous Driving. (80%)
Self-supervised Adversarial Training of Monocular Depth Estimation against Physical-World Attacks. (67%)
SlowPerception: Physical-World Latency Attack against Visual Perception in Autonomous Driving. (64%)
ProFeAT: Projected Feature Adversarial Training for Self-Supervised Learning of Robust Representations. (38%)
Certified Robustness to Data Poisoning in Gradient-Based Training. (22%)
Machine Against the RAG: Jamming Retrieval-Augmented Generation with Blocker Documents. (4%)
PSBD: Prediction Shift Uncertainty Unlocks Backdoor Detection. (2%)
A Relevance Model for Threat-Centric Ranking of Cybersecurity Vulnerabilities. (1%)
Safety Alignment Should Be Made More Than Just a Few Tokens Deep. (1%)
Injecting Undetectable Backdoors in Obfuscated Neural Networks and Language Models. (1%)
2024-06-08
SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner. (99%)
One Perturbation is Enough: On Generating Universal Adversarial Perturbations against Vision-Language Pre-training Models. (99%)
Bridging the Gap: Rademacher Complexity in Robust and Standard Generalization. (98%)
Perturbation Towards Easy Samples Improves Targeted Adversarial Transferability. (96%)
Enhancing Adversarial Transferability via Information Bottleneck Constraints. (68%)
Exploring Adversarial Robustness of Deep State Space Models. (56%)
Adversarial flows: A gradient flow characterization of adversarial attacks. (13%)
2024-06-07
ADBA:Approximation Decision Boundary Approach for Black-Box Adversarial Attacks. (99%)
Probabilistic Perspectives on Error Minimization in Adversarial Reinforcement Learning. (98%)
Corpus Poisoning via Approximate Greedy Gradient Descent. (86%)
Compositional Curvature Bounds for Deep Neural Networks. (84%)
Adversarial Tuning: Defending Against Jailbreak Attacks for LLMs. (41%)
Clarifying Myths About the Relationship Between Shape Bias, Accuracy, and Robustness. (22%)
GENIE: Watermarking Graph Neural Networks for Link Prediction. (15%)
The Price of Implicit Bias in Adversarially Robust Generalization. (5%)
Contextual fusion enhances robustness to image blurring. (5%)
LLM Whisperer: An Inconspicuous Attack to Bias LLM Responses. (1%)
2024-06-06
Batch-in-Batch: a new adversarial training framework for initial perturbation and sample selection. (99%)
Talos: A More Effective and Efficient Adversarial Defense for GNN Models Based on the Global Homophily of Graphs. (98%)
Improving Alignment and Robustness with Circuit Breakers. (98%)
Behavior-Targeted Attack on Reinforcement Learning with Limited Access to Victim's Policy. (76%)
AutoJailbreak: Exploring Jailbreak Attacks and Defenses through a Dependency Lens. (69%)
Neural Codec-based Adversarial Sample Detection for Speaker Verification. (68%)
Interpreting the Second-Order Effects of Neurons in CLIP. (67%)
Jailbreak Vision Language Models via Bi-Modal Adversarial Prompt. (56%)
Memorization in deep learning: A survey. (1%)
2024-06-05
ZeroPur: Succinct Training-Free Adversarial Purification. (99%)
VQUNet: Vector Quantization U-Net for Defending Adversarial Atacks by Regularizing Unwanted Noise. (99%)
DifAttack++: Query-Efficient Black-Box Adversarial Attack via Hierarchical Disentangled Feature Space in Cross-Domain. (99%)
Distributional Adversarial Loss. (96%)
Defending Large Language Models Against Attacks With Residual Stream Activation Analysis. (83%)
Graph Neural Network Explanations are Fragile. (80%)
A Geometric View of Data Complexity: Efficient Local Intrinsic Dimension Estimation with Diffusion Models. (68%)
Principles of Designing Robust Remote Face Anti-Spoofing Systems. (13%)
Mutual Information Guided Backdoor Mitigation for Pre-trained Encoders. (13%)
JIGMARK: A Black-Box Approach for Enhancing Image Watermarks against Diffusion Model Edits. (10%)
Are Your Models Still Fair? Fairness Attacks on Graph Neural Networks via Node Injections. (10%)
Enhancing the Resilience of Graph Neural Networks to Topological Perturbations in Sparse Graphs. (8%)
Reconstructing training data from document understanding models. (1%)
FREA: Feasibility-Guided Generation of Safety-Critical Scenarios with Reasonable Adversariality. (1%)
2024-06-04
Advancing Generalized Transfer Attack with Initialization Derived Bilevel Optimization and Dynamic Sequence Truncation. (99%)
Effects of Exponential Gaussian Distribution on (Double Sampling) Randomized Smoothing. (98%)
PuFace: Defending against Facial Cloaking Attacks for Facial Recognition Models. (54%)
A Risk Estimation Study of Native Code Vulnerabilities in Android Applications. (5%)
Verifying the Generalization of Deep Learning to Out-of-Distribution Domains. (3%)
Large Language Models as Carriers of Hidden Messages. (2%)
Nonlinear Transformations Against Unlearnable Datasets. (2%)
Inference Attacks: A Taxonomy, Survey, and Promising Directions. (1%)
QROA: A Black-Box Query-Response Optimization Attack on LLMs. (1%)
The Crystal Ball Hypothesis in diffusion models: Anticipating object positions from initial noise. (1%)
Can Dense Connectivity Benefit Outlier Detection? An Odyssey with NAS. (1%)
2024-06-03
Constraint-based Adversarial Example Synthesis. (99%)
SVASTIN: Sparse Video Adversarial Attack via Spatio-Temporal Invertible Neural Networks. (99%)
Reproducibility Study on Adversarial Attacks Against Robust Transformer Trackers. (93%)
CR-UTP: Certified Robustness against Universal Text Perturbations on Large Language Models. (83%)
Are AI-Generated Text Detectors Robust to Adversarial Perturbations? (80%)
Model for Peanuts: Hijacking ML Models without Training Access is Possible. (62%)
SLANT: Spurious Logo ANalysis Toolkit. (47%)
MedFuzz: Exploring the Robustness of Large Language Models in Medical Question Answering. (16%)
From Feature Visualization to Visual Circuits: Effect of Adversarial Model Manipulation. (12%)
A Game-Theoretic Approach to Privacy-Utility Tradeoff in Sharing Genomic Summary Statistics. (10%)
Poisoning Attacks and Defenses in Recommender Systems: A Survey. (10%)
Unelicitable Backdoors in Language Models via Cryptographic Transformer Circuits. (4%)
PRICE: A Pretrained Model for Cross-Database Cardinality Estimation. (1%)
2024-06-02
Constrained Adaptive Attack: Effective Adversarial Attack Against Deep Neural Networks for Tabular Data. (99%)
Improving Accuracy-robustness Trade-off via Pixel Reweighted Adversarial Training. (98%)
A Novel Defense Against Poisoning Attacks on Federated Learning: LayerCAM Augmented with Autoencoder. (31%)
Towards General Robustness Verification of MaxPool-based Convolutional Neural Networks via Tightening Linear Approximation. (13%)
Assessing the Adversarial Security of Perceptual Hashing Algorithms. (9%)
Invisible Backdoor Attacks on Diffusion Models. (2%)
2024-06-01
Robust Knowledge Distillation Based on Feature Variance Against Backdoored Teacher Model. (3%)
2024-05-31
Query Provenance Analysis: Efficient and Robust Defense against Query-based Black-box Attacks. (99%)
Investigating and unmasking feature-level vulnerabilities of CNNs to adversarial perturbations. (95%)
Robust Stable Spiking Neural Networks. (38%)
Improved Techniques for Optimization-Based Jailbreaking on Large Language Models. (26%)
ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learning. (22%)
Enhancing Jailbreak Attack Against Large Language Models through Silent Tokens. (13%)
Exploring Vulnerabilities and Protections in Large Language Models: A Survey. (10%)
GANcrop: A Contrastive Defense Against Backdoor Attacks in Federated Learning. (9%)
Neural Network Verification with Branch-and-Bound for General Nonlinearities. (9%)
StyDeSty: Min-Max Stylization and Destylization for Single Domain Generalization. (4%)
GI-NAS: Boosting Gradient Inversion Attacks through Adaptive Neural Architecture Search. (1%)
2024-05-30
Disrupting Diffusion: Token-Level Attention Erasure Attack against Diffusion-based Customization. (99%)
HOLMES: to Detect Adversarial Examples with Multiple Detectors. (99%)
Typography Leads Semantic Diversifying: Amplifying Adversarial Transferability across Multimodal Large Language Models. (99%)
Enhancing Adversarial Robustness in SNNs with Sparse Gradients. (92%)
Exploring the Robustness of Decision-Level Through Adversarial Attacks on LLM-Based Embodied Models. (89%)
Phantom: General Trigger Attacks on Retrieval Augmented Language Generation. (83%)
SleeperNets: Universal Backdoor Poisoning Attacks Against Reinforcement Learning Agents. (75%)
Deep Learning Approaches for Detecting Adversarial Cyberbullying and Hate Speech in Social Networks. (73%)
BAN: Detecting Backdoors Activated by Adversarial Neuron Noise. (68%)
Is My Data in Your Retrieval Database? Membership Inference Attacks Against Retrieval Augmented Generation. (45%)
Defensive Prompt Patch: A Robust and Interpretable Defense of LLMs against Jailbreak Attacks. (38%)
Large Language Model Watermark Stealing With Mixed Integer Programming. (33%)
DiffPhysBA: Diffusion-based Physical Backdoor Attack against Person Re-Identification in Real-World. (22%)
Investigating the Robustness of LLMs on Math Word Problems. (16%)
Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness. (5%)
Certifying Global Robustness for Deep Neural Networks. (2%)
Breaking Indistinguishability with Transfer Learning: A First Look at SPECK32/64 Lightweight Block Ciphers. (1%)
Reconstruction Attacks on Machine Unlearning: Simple Models are Vulnerable. (1%)
2024-05-29
Efficient Black-box Adversarial Attacks via Bayesian Optimization Guided by a Function Prior. (99%)
Leveraging Many-To-Many Relationships for Defending Against Visual-Language Adversarial Attacks. (96%)
Model Agnostic Defense against Adversarial Patch Attacks on Object Detection in Unmanned Aerial Vehicles. (92%)
Diffusion Policy Attacker: Crafting Adversarial Attacks for Diffusion-based Policies. (92%)
Evaluating the Effectiveness and Robustness of Visual Similarity-based Phishing Detection Models. (91%)
Verifiably Robust Conformal Prediction. (82%)
AI Risk Management Should Incorporate Both Safety and Security. (67%)
AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided Optimization. (61%)
EntProp: High Entropy Propagation for Improving Accuracy and Robustness. (50%)
ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruning. (26%)
Resurrecting Old Classes with New Data for Exemplar-Free Continual Learning. (22%)
Node Injection Attack Based on Label Propagation Against Graph Neural Network. (12%)
Genshin: General Shield for Natural Language Processing with Large Language Models. (5%)
Confronting the Reproducibility Crisis: A Case Study of Challenges in Cybersecurity AI. (2%)
Enhancing Security and Privacy in Federated Learning using Update Digests and Voting-Based Defense. (1%)
Gone but Not Forgotten: Improved Benchmarks for Machine Unlearning. (1%)
MemControl: Mitigating Memorization in Diffusion Models via Automated Parameter Selection. (1%)
2024-05-28
Towards Unified Robustness Against Both Backdoor and Adversarial Attacks. (99%)
Improved Generation of Adversarial Examples Against Safety-aligned LLMs. (99%)
PureGen: Universal Data Purification for Train-Time Poison Defense via Generative Model Dynamics. (98%)
PureEBM: Universal Poison Purification via Mid-Run Dynamics of Energy-Based Models. (98%)
White-box Multimodal Jailbreaks Against Large Vision-Language Models. (96%)
Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editing. (92%)
Wavelet-Based Image Tokenizer for Vision Transformers. (64%)
Cross-Context Backdoor Attacks against Graph Prompt Learning. (13%)
BlueSWAT: A Lightweight State-Aware Security Framework for Bluetooth Low Energy. (1%)
Watermarking Counterfactual Explanations. (1%)
Black-Box Detection of Language Model Watermarks. (1%)
2024-05-27
Adversarial Attacks on Both Face Recognition and Face Anti-spoofing Models. (99%)
The Uncanny Valley: Exploring Adversarial Robustness from a Flatness Perspective. (99%)
Exploiting the Layered Intrinsic Dimensionality of Deep Models for Practical Adversarial Training. (98%)
Spectral regularization for adversarially-robust representation learning. (86%)
TIMA: Text-Image Mutual Awareness for Balancing Zero-Shot Adversarial Robustness and Generalization Ability. (83%)
OSLO: One-Shot Label-Only Membership Inference Attacks. (81%)
Verifying Properties of Binary Neural Networks Using Sparse Polynomial Optimization. (33%)
Rethinking Pruning for Backdoor Mitigation: An Optimization Perspective. (26%)
Navigating the Safety Landscape: Measuring Risks in Finetuning Large Language Models. (8%)
Can We Trust Embodied Agents? Exploring Backdoor Attacks against Embodied LLM-based Decision-Making Systems. (5%)
LabObf: A Label Protection Scheme for Vertical Federated Learning Through Label Obfuscation. (1%)
Magnitude-based Neuron Pruning for Backdoor Defens. (1%)
2024-05-26
Medical MLLM is Vulnerable: Cross-Modality Jailbreak and Mismatched Attacks on Medical Multimodal Large Language Models. (67%)
Pruning for Robust Concept Erasing in Diffusion Models. (38%)
TrojFM: Resource-efficient Backdoor Attacks against Very Large Foundation Models. (31%)
Partial train and isolate, mitigate backdoor attack. (1%)
Automatic Jailbreaking of the Text-to-Image Generative AI Systems. (1%)
2024-05-25
Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attack. (99%)
Detecting Adversarial Data via Perturbation Forgery. (99%)
Enhancing Adversarial Transferability Through Neighborhood Conditional Sampling. (98%)
R.A.C.E.: Robust Adversarial Concept Erasure for Secure Text-to-Image Diffusion Model. (97%)
Uncertainty Measurement of Deep Learning System based on the Convex Hull of Training Sets. (89%)
Layer-Aware Analysis of Catastrophic Overfitting: Revealing the Pseudo-Robust Shortcut Dependency. (81%)
Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor. (70%)
Visual-RolePlay: Universal Jailbreak Attack on MultiModal Large Language Models via Role-playing Image Character. (56%)
Intruding with Words: Towards Understanding Graph Injection Attacks at the Text Level. (8%)
No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks. (4%)
Robust Message Embedding via Attention Flow-Based Steganography. (1%)
2024-05-24
Robust width: A lightweight and certifiable adversarial defense. (99%)
Large Language Model Sentinel: LLM Agent for Adversarial Purification. (99%)
Adversarial Attacks on Hidden Tasks in Multi-Task Learning. (98%)
Evaluating and Safeguarding the Adversarial Robustness of Retrieval-Based In-Context Learning. (95%)
Certifying Adapters: Enabling and Enhancing the Certification of Classifier Adversarial Robustness. (92%)
Efficient Adversarial Training in LLMs with Continuous Attacks. (92%)
Rethinking Independent Cross-Entropy Loss For Graph-Structured Data. (76%)
BDetCLIP: Multimodal Prompting Contrastive Test-Time Backdoor Detection. (61%)
Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models. (47%)
Robustifying Safety-Aligned Large Language Models through Clean Data Curation. (15%)
HiddenSpeaker: Generate Imperceptible Unlearnable Audios for Speaker Verification System. (15%)
Can Implicit Bias Imply Adversarial Robustness? (11%)
Certifiably Robust RAG against Retrieval Corruption. (10%)
BadGD: A unified data-centric framework to identify gradient descent vulnerabilities. (8%)
AuthNet: Neural Network with Integrated Authentication Logic. (5%)
Revisit, Extend, and Enhance Hessian-Free Influence Functions. (2%)
2024-05-23
Eidos: Efficient, Imperceptible Adversarial 3D Point Clouds. (98%)
Certified Robustness against Sparse Adversarial Perturbations via Data Localization. (92%)
A New Formulation for Zeroth-Order Optimization of Adversarial EXEmples in Malware Detection. (91%)
Generating camera failures as a class of physics-based adversarial examples. (87%)
TrojanForge: Generating Adversarial Hardware Trojan Examples with Reinforcement Learning. (84%)
Towards Transferable Attacks Against Vision-LLMs in Autonomous Driving with Typography. (83%)
SLIFER: Investigating Performance and Robustness of Malware Detection Pipelines. (81%)
How Does Bayes Error Limit Probabilistic Robust Accuracy. (76%)
Universal Robustness via Median Randomized Smoothing for Real-World Super-Resolution. (67%)
Towards Imperceptible Backdoor Attack in Self-supervised Learning. (61%)
Unveiling the Achilles' Heel of NLG Evaluators: A Unified Adversarial Framework Driven by Large Language Models. (33%)
AdjointDEIS: Efficient Gradients for Diffusion Models. (15%)
What Variables Affect Out-of-Distribution Generalization in Pretrained Models? (9%)
Tighter Privacy Auditing of DP-SGD in the Hidden State Threat Model. (8%)
RFLPA: A Robust Federated Learning Framework against Poisoning Attacks with Secure Aggregation. (1%)
Are You Copying My Prompt? Protecting the Copyright of Vision Prompt for VPaaS via Watermark. (1%)
2024-05-22
Learning to Transform Dynamically for Better Adversarial Transferability. (99%)
Adversarial Training of Two-Layer Polynomial and ReLU Activation Networks via Convex Optimization. (80%)
Towards Certification of Uncertainty Calibration under Adversarial Attacks. (75%)
LookHere: Vision Transformers with Directed Attention Generalize and Extrapolate. (67%)
Remote Keylogging Attacks in Multi-user VR Applications. (13%)
Nearly Tight Black-Box Auditing of Differentially Private Machine Learning. (5%)
WordGame: Efficient & Effective LLM Jailbreak via Simultaneous Obfuscation in Query and Response. (1%)
2024-05-21
Mellivora Capensis: A Backdoor-Free Training Framework on the Poisoned Dataset without Auxiliary Data. (92%)
Adversarial Training via Adaptive Knowledge Amalgamation of an Ensemble of Teachers. (87%)
Rethinking the Vulnerabilities of Face Recognition Systems:From a Practical Perspective. (78%)
EmInspector: Combating Backdoor Attacks in Federated Self-Supervised Learning Through Embedding Inspection. (47%)
Fully Randomized Pointers. (15%)
A novel reliability attack of Physical Unclonable Functions. (13%)
Nearest is Not Dearest: Towards Practical Defense against Quantization-conditioned Backdoor Attacks. (8%)
Dullahan: Stealthy Backdoor Attack against Without-Label-Sharing Split Learning. (4%)
Tiny Refinements Elicit Resilience: Toward Efficient Prefix-Model Against LLM Red-Teaming. (1%)
2024-05-20
A Constraint-Enforcing Reward for Adversarial Attacks on Text Classifiers. (99%)
GAN-GRID: A Novel Generative Attack on Smart Grid Stability Prediction. (98%)
Robust Deep Reinforcement Learning with Adaptive Adversarial Perturbations in Action Space. (76%)
EGAN: Evolutional GAN for Ransomware Evasion. (74%)
Rethinking Robustness Assessment: Adversarial Attacks on Learning-based Quadrupedal Locomotion Controllers. (31%)
Adversarially Diversified Rehearsal Memory (ADRM): Mitigating Memory Overfitting Challenge in Continual Learning. (8%)
Efficient Model-Stealing Attacks Against Inductive Graph Neural Networks. (3%)
DispaRisk: Auditing Fairness Through Usable Information. (1%)
2024-05-19
Adaptive Batch Normalization Networks for Adversarial Robustness. (99%)
An Invisible Backdoor Attack Based On Semantic Feature. (96%)
Certified Robust Accuracy of Neural Networks Are Bounded due to Bayes Errors. (81%)
A GAN-Based Data Poisoning Attack Against Federated Learning Systems and Its Countermeasure. (68%)
SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks. (62%)
Fed-Credit: Robust Federated Learning with Credibility Management. (13%)
BOSC: A Backdoor-based Framework for Open Set Synthetic Image Attribution. (5%)
2024-05-18
Towards Robust Policy: Enhancing Offline Reinforcement Learning with Adversarial Attacks and Defenses. (84%)
Trustworthy Actionable Perturbations. (82%)
Fully Exploiting Every Real Sample: SuperPixel Sample Gradient Model Stealing. (13%)
UPAM: Unified Prompt Attack in Text-to-Image Generation Models Against Both Textual Filters and Visual Checkers. (12%)
BadActs: A Universal Backdoor Defense in the Activation Space. (10%)
On Robust Reinforcement Learning with Lipschitz-Bounded Policy Networks. (8%)
Diffusion Model Driven Test-Time Image Adaptation for Robust Skin Lesion Classification. (3%)
2024-05-17
Revisiting the Robust Generalization of Adversarial Prompt Tuning. (99%)
Safeguarding Vision-Language Models Against Patched Visual Prompt Injectors. (99%)
Rethinking Graph Backdoor Attacks: A Distribution-Preserving Perspective. (83%)
Not All Prompts Are Secure: A Switchable Backdoor Attack Against Pre-trained Vision Transformers. (67%)
Boosting Few-Pixel Robustness Verification via Covering Verification Designs. (1%)
2024-05-16
DiffAM: Diffusion-based Adversarial Makeup Transfer for Facial Privacy Protection. (99%)
Infrared Adversarial Car Stickers. (98%)
Adversarial Robustness for Visual Grounding of Multimodal Large Language Models. (95%)
Adversarial Robustness Guarantees for Quantum Classifiers. (81%)
Box-Free Model Watermarks Are Prone to Black-Box Removal Attacks. (13%)
Relational DNN Verification With Cross Executional Bound Refinement. (8%)
Manifold Integrated Gradients: Riemannian Geometry for Feature Attribution. (1%)
Dealing Doubt: Unveiling Threat Models in Gradient Inversion Attacks under Federated Learning, A Survey and Taxonomy. (1%)
2024-05-15
Properties that allow or prohibit transferability of adversarial attacks among quantized networks. (99%)
Towards Evaluating the Robustness of Automatic Speech Recognition Systems via Audio Style Transfer. (99%)
Cross-Input Certified Training for Universal Perturbations. (98%)
IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency. (4%)
Themis: Automatic and Efficient Deep Learning System Testing with Strong Fault Detection Capability. (4%)
Optimizing Sensor Network Design for Multiple Coverage. (1%)
2024-05-14
SpeechGuard: Exploring the Adversarial Robustness of Multimodal Large Language Models. (99%)
Certifying Robustness of Graph Convolutional Networks for Node Perturbation with Polyhedra Abstract Interpretation. (92%)
The Pitfalls and Promise of Conformal Inference Under Adversarial Attacks. (92%)
The RoboDrive Challenge: Drive Anytime Anywhere in Any Condition. (11%)
Pointwise Lipschitz Continuous Graph Algorithms via Proximal Gradient Analysis. (1%)
Achieving Resolution-Agnostic DNN-based Image Watermarking:A Novel Perspective of Implicit Neural Representation. (1%)
Neural Collapse Meets Differential Privacy: Curious Behaviors of NoisyGD with Near-perfect Representation Learning. (1%)
UnMarker: A Universal Attack on Defensive Watermarking. (1%)
RS-Reg: Probabilistic and Robust Certified Regression Through Randomized Smoothing. (1%)
2024-05-13
Environmental Matching Attack Against Unmanned Aerial Vehicles Object Detection. (96%)
CrossCert: A Cross-Checking Detection Approach to Patch Robustness Certification for Deep Learning Models. (82%)
RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectors. (15%)
GLiRA: Black-Box Membership Inference Attack via Knowledge Distillation. (11%)
Backdoor Removal for Generative Large Language Models. (1%)
2024-05-11
Stealthy Imitation: Reward-guided Environment-free Policy Stealing. (1%)
2024-05-10
Improving Transferable Targeted Adversarial Attack via Normalized Logit Calibration and Truncated Feature Mixing. (99%)
Disttack: Graph Adversarial Attacks Toward Distributed GNN Training. (98%)
Exploring the Interplay of Interpretability and Robustness in Deep Neural Networks: A Saliency-guided Approach. (98%)
Evaluating Adversarial Robustness in the Spatial Frequency Domain. (96%)
Certified $\ell_2$ Attribution Robustness via Uniformly Smoothed Attributions. (96%)
PUMA: margin-based data pruning. (80%)
2024-05-09
BB-Patch: BlackBox Adversarial Patch-Attack using Zeroth-Order Optimization. (99%)
Muting Whisper: A Universal Acoustic Adversarial Attack on Speech Foundation Models. (97%)
Poisoning-based Backdoor Attacks for Arbitrary Target Label with Positive Triggers. (80%)
Link Stealing Attacks Against Inductive Graph Neural Networks. (75%)
Hard Work Does Not Always Pay Off: Poisoning Attacks on Neural Architecture Search. (68%)
Concealing Backdoor Model Updates in Federated Learning by Trigger-Optimized Data Poisoning. (62%)
Towards Robust Physical-world Backdoor Attacks on Lane Detection. (50%)
Model Inversion Robustness: Can Transfer Learning Help? (45%)
Chain of Attack: a Semantic-Driven Contextual Multi-Turn attacker for LLM. (3%)
Demystifying Behavior-Based Malware Detection at Endpoints. (2%)
2024-05-08
Universal Adversarial Perturbations for Vision-Language Pre-trained Models. (99%)
Adversarial Threats to Automatic Modulation Open Set Recognition in Wireless Networks. (99%)
Untargeted Adversarial Attack on Knowledge Graph Embeddings. (98%)
Towards Efficient Training and Evaluation of Robust Models against $l_0$ Bounded Adversarial Perturbations. (98%)
Towards Accurate and Robust Architectures via Neural Architecture Search. (96%)
Explanation as a Watermark: Towards Harmless and Multi-bit Model Ownership Verification via Watermarking Feature Attribution. (1%)
2024-05-07
Revisiting character-level adversarial attacks. (99%)
Explainability-Informed Targeted Malware Misclassification. (99%)
Effective and Robust Adversarial Training against Data and Label Corruptions. (70%)
Going Proactive and Explanatory Against Malware Concept Drift. (1%)
Verified Neural Compressed Sensing. (1%)
2024-05-06
Exploring Frequencies via Feature Mixing and Meta-Learning for Improving Adversarial Transferability. (99%)
Cutting through buggy adversarial example defenses: fixing 1 line of code breaks Sabre. (99%)
On Adversarial Examples for Text Classification by Perturbing Latent Representations. (99%)
Is ReLU Adversarially Robust? (98%)
Enhancing O-RAN Security: Evasion Attacks and Robust Defenses for Graph Reinforcement Learning-based Connection Management. (91%)
BadFusion: 2D-Oriented Backdoor Attacks against 3D Object Detection. (75%)
Provably Unlearnable Data Examples. (64%)
DarkFed: A Data-Free Backdoor Attack in Federated Learning. (33%)
UnsafeBench: Benchmarking Image Safety Classifiers on Real-World and AI-Generated Images. (1%)
Why is SAM Robust to Label Noise? (1%)
Detecting Android Malware: From Neural Embeddings to Hands-On Validation with BERTroid. (1%)
LaserEscape: Detecting and Mitigating Optical Probing Attacks. (1%)
2024-05-05
Defense against Joint Poison and Evasion Attacks: A Case Study of DERMS. (88%)
To Each (Textual Sequence) Its Own: Improving Memorized-Data Unlearning in Large Language Models. (15%)
Explainable Malware Detection with Tailored Logic Explained Networks. (2%)
2024-05-04
Leveraging the Human Ventral Visual Stream to Improve Neural Network Robustness. (92%)
Updating Windows Malware Detectors: Balancing Robustness and Regression against Adversarial EXEmples. (83%)
Assessing Adversarial Robustness of Large Language Models: An Empirical Study. (76%)
2024-05-03
A Novel Approach to Guard from Adversarial Attacks using Stable Diffusion. (99%)
From Attack to Defense: Insights into Deep Learning Security Measures in Black-Box Settings. (99%)
ProFLingo: A Fingerprinting-based Copyright Protection Scheme for Large Language Models. (97%)
Impact of Architectural Modifications on Deep Learning Adversarial Robustness. (88%)
Adaptive and robust watermark against model extraction attack. (38%)
Robust Explainable Recommendation. (9%)
Adversarial Botometer: Adversarial Analysis for Social Bot Detection. (1%)
2024-05-02
Position Paper: Beyond Robustness Against Single Attack Types. (99%)
Explainability Guided Adversarial Evasion Attacks on Malware Detectors. (98%)
Purify Unlearnable Examples via Rate-Constrained Variational Autoencoders. (88%)
Poisoning Attacks on Federated Learning for Autonomous Driving. (75%)
Adversarial Attacks on Reinforcement Learning Agents for Command and Control. (75%)
Boosting Jailbreak Attack with Momentum. (41%)
Uniformly Stable Algorithms for Adversarial Training and Beyond. (10%)
ATTAXONOMY: Unpacking Differential Privacy Guarantees Against Practical Adversaries. (2%)
2024-05-01
Certified Adversarial Robustness of Machine Learning-based Malware Detectors via (De)Randomized Smoothing. (99%)
JNI Global References Are Still Vulnerable: Attacks and Defenses. (12%)
Robustness of graph embedding methods for community detection. (2%)
Exploiting Positional Bias for Query-Agnostic Generative Content in Search. (1%)
2024-04-30
Revisiting the Adversarial Robustness of Vision Language Models: a Multimodal Perspective. (99%)
Probing Unlearned Diffusion Models: A Transferable Adversarial Attack Perspective. (99%)
AttackBench: Evaluating Gradient-based Attacks for Adversarial Examples. (99%)
Provably Robust Conformal Prediction with Improved Efficiency. (98%)
ASAM: Boosting Segment Anything Model with Adversarial Tuning. (98%)
Adversarial Attacks and Defense for Conversation Entailment Task. (98%)
Causal Perception Inspired Representation Learning for Trustworthy Image Quality Assessment. (92%)
Transferring Troubles: Cross-Lingual Transferability of Backdoor Attacks in LLMs with Instruction Tuning. (81%)
Let's Focus: Focused Backdoor Attack against Federated Transfer Learning. (75%)
VeriFence: Lightweight and Precise Spectre Defenses for Untrusted Linux Kernel Extensions. (1%)
Physical Backdoor: Towards Temperature-based Backdoor Attacks in the Physical World. (1%)
2024-04-29
Assessing Cybersecurity Vulnerabilities in Code Large Language Models. (99%)
A Systematic Evaluation of Adversarial Attacks against Speech Emotion Recognition Models. (99%)
Certification of Speaker Recognition Models to Additive Perturbations. (54%)
Espresso: Robust Concept Filtering in Text-to-Image Models. (15%)
Why You Should Not Trust Interpretations in Machine Learning: Adversarial Attacks on Partial Dependence Plots. (13%)
Machine Learning for Windows Malware Detection and Classification: Methods, Challenges and Ongoing Research. (3%)
Towards Quantitative Evaluation of Explainable AI Methods for Deepfake Detection. (1%)
Harmonic Machine Learning Models are Robust. (1%)
Enhancing IoT Security: A Novel Feature Engineering Approach for ML-Based Intrusion Detection Systems. (1%)
2024-04-28
Learnable Linguistic Watermarks for Tracing Model Extraction Attacks on Large Language Models. (1%)
2024-04-27
Towards Robust Recommendation: A Review and an Adversarial Robustness Evaluation Library. (92%)
Privacy-Preserving Aggregation for Decentralized Learning with Byzantine-Robustness. (70%)
Bounding the Expected Robustness of Graph Neural Networks Subject to Node Feature Attacks. (67%)
Are Watermarks Bugs for Deepfake Detectors? Rethinking Proactive Forensics. (2%)
2024-04-26
Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networks. (99%)
Adversarial Examples: Generation Proposal in the Context of Facial Recognition Systems. (92%)
Human-Imperceptible Retrieval Poisoning Attacks in LLM-Powered Applications. (54%)
Evaluations of Machine Learning Privacy Defenses are Misleading. (3%)
Enhancing Privacy and Security of Autonomous UAV Navigation. (2%)
Adversarial Reweighting with $\alpha$-Power Maximization for Domain Adaptation. (1%)
Changing the Training Data Distribution to Reduce Simplicity Bias Improves In-distribution Generalization. (1%)
Adversarial Consistency and the Uniqueness of the Adversarial Bayes Classifier. (1%)
2024-04-25
Generating Minimalist Adversarial Perturbations to Test Object-Detection Models: An Adaptive Multi-Metric Evolutionary Search Approach. (98%)
PAD: Patch-Agnostic Defense against Adversarial Patch Attacks. (92%)
Defending Spiking Neural Networks against Adversarial Attacks through Image Purification. (84%)
Don't Say No: Jailbreaking LLM by Suppressing Refusal. (67%)
A Self-Organizing Clustering System for Unsupervised Distribution Shift Detection. (12%)
Constructing Optimal Noise Channels for Enhanced Robustness in Quantum Machine Learning. (2%)
Energy-Latency Manipulation of Multi-modal Large Language Models via Verbose Samples. (2%)
Talking Nonsense: Probing Large Language Models' Understanding of Adversarial Gibberish Inputs. (1%)
2024-04-24
Steal Now and Attack Later: Evaluating Robustness of Object Detection against Black-box Adversarial Attacks. (99%)
An Analysis of Recent Advances in Deepfake Image Detection in an Evolving Threat Landscape. (99%)
An Empirical Study of Aegis. (98%)
A General Black-box Adversarial Attack on Graph-based Fake News Detectors. (96%)
A Comparative Analysis of Adversarial Robustness for Quantum and Classical Machine Learning Models. (83%)
MISLEAD: Manipulating Importance of Selected features for Learning Epsilon in Evasion Attack Deception. (83%)
Investigating the prompt leakage effect and black-box defenses for multi-turn LLM interactions. (45%)
Universal Adversarial Triggers Are Not Universal. (16%)
CLAD: Robust Audio Deepfake Detection Against Manipulation Attacks with Contrastive Learning. (2%)
2024-04-23
Security Analysis of WiFi-based Sensing Systems: Threats from Perturbation Attacks. (61%)
Manipulating Recommender Systems: A Survey of Poisoning Attacks and Countermeasures. (61%)
PoisonedFL: Model Poisoning Attacks to Federated Learning via Multi-Round Consistency. (54%)
Perturbing Attention Gives You More Bang for the Buck: Subtle Imaging Perturbations That Efficiently Fool Customized Diffusion Models. (47%)
Talk Too Much: Poisoning Large Language Models under Token Limit. (38%)
Leverage Variational Graph Representation For Model Poisoning on Federated Learning. (10%)
Formal Verification of Graph Convolutional Networks with Uncertain Node Features and Uncertain Graph Structure. (2%)
Does It Make Sense to Explain a Black Box With Another Black Box? (1%)
Graph Machine Learning in the Era of Large Language Models (LLMs). (1%)
2024-04-22
Towards Understanding the Robustness of Diffusion-Based Purification: A Stochastic Perspective. (98%)
Double Privacy Guard: Robust Traceable Adversarial Watermarking against Face Recognition. (93%)
CloudFort: Enhancing Robustness of 3D Point Cloud Classification Against Backdoor Attacks via Spatial Partitioning and Ensemble Prediction. (74%)
Explicit Lipschitz Value Estimation Enhances Policy Robustness Against Perturbation. (67%)
Audio Anti-Spoofing Detection: A Survey. (62%)
Dual Model Replacement:invisible Multi-target Backdoor Attack based on Federal Learning. (41%)
Protecting Your LLMs with Information Bottleneck. (26%)
Competition Report: Finding Universal Jailbreak Backdoors in Aligned LLMs. (13%)
Deep Learning as Ricci Flow. (2%)
Hyp-OC: Hyperbolic One Class Classification for Face Anti-Spoofing. (1%)
Poisoning Attacks on Federated Learning-based Wireless Traffic Prediction. (1%)
Typos that Broke the RAG's Back: Genetic Attack on RAG Pipeline by Simulating Documents in the Wild via Low-level Perturbations. (1%)
2024-04-21
Attack on Scene Flow using Point Clouds. (98%)
Fermi-Bose Machine. (96%)
Robust EEG-based Emotion Recognition Using an Inception and Two-sided Perturbation Model. (50%)
AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs. (47%)
Swap It Like Its Hot: Segmentation-based spoof attacks on eye-tracking images. (26%)
Trojan Detection in Large Language Models: Insights from The Trojan Detection Challenge. (1%)
2024-04-20
Reliable Model Watermarking: Defending Against Theft without Compromising on Evasion. (99%)
Beyond Score Changes: Adversarial Attack on No-Reference Image Quality Assessment from Two Perspectives. (99%)
Pixel is a Barrier: Diffusion Models Are More Adversarially Robust Than We Think. (99%)
Backdoor Attacks and Defenses on Semantic-Symbol Reconstruction in Semantic Communications. (41%)
2024-04-19
How Real Is Real? A Human Evaluation Framework for Unrestricted Adversarial Examples. (99%)
AED-PADA:Improving Generalizability of Adversarial Example Detection via Principal Adversarial Domain Adaptation. (99%)
A Clean-graph Backdoor Attack against Graph Convolutional Networks with Poisoned Label Only. (75%)
Physical Backdoor Attack can Jeopardize Driving with Vision-Large-Language Models. (5%)
MLSD-GAN -- Generating Strong High Quality Face Morphing Attacks using Latent Semantic Disentanglement. (3%)
Model-Based Counterfactual Explanations Incorporating Feature Space Attributes for Tabular Data. (1%)
LSP Framework: A Compensatory Model for Defeating Trigger Reverse Engineering via Label Smoothing Poisoning. (1%)
2024-04-18
Fortify the Guardian, Not the Treasure: Resilient Adversarial Detectors. (99%)
Advancing the Robustness of Large Language Models through Self-Denoised Smoothing. (98%)
SA-Attack: Speed-adaptive stealthy adversarial attack on trajectory prediction. (98%)
Enhance Robustness of Language Models Against Variation Attack through Graph Integration. (33%)
Uncovering Safety Risks of Large Language Models through Concept Activation Vector. (22%)
Proteus: Preserving Model Confidentiality during Graph Optimizations. (15%)
Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models. (2%)
Is There No Such Thing as a Bad Question? H4R: HalluciBot For Ratiocination, Rewriting, Ranking, and Routing. (1%)
2024-04-17
The Victim and The Beneficiary: Exploiting a Poisoned Model to Train a Clean Model on Poisoned Data. (83%)
GenFighter: A Generative and Evolutive Textual Attack Removal. (82%)
Utilizing Adversarial Examples for Bias Mitigation and Accuracy Enhancement. (80%)
Exploring DNN Robustness Against Adversarial Attacks Using Approximate Multipliers. (75%)
Exploring the Transferability of Visual Prompting for Multimodal Large Language Models. (2%)
Toward Understanding the Disagreement Problem in Neural Network Feature Attribution. (1%)
Detector Collapse: Backdooring Object Detection to Catastrophic Overload or Blindness. (1%)
Towards Robust and Interpretable EMG-based Hand Gesture Recognition using Deep Metric Meta Learning. (1%)
2024-04-16
Efficiently Adversarial Examples Generation for Visual-Language Models under Targeted Transfer Scenarios using Diffusion Models. (99%)
Adversarial Identity Injection for Semantic Face Image Synthesis. (38%)
Robust Noisy Label Learning via Two-Stream Sample Distillation. (1%)
2024-04-15
Black-box Adversarial Transferability: An Empirical Study in Cybersecurity Perspective. (99%)
Towards a Novel Perspective on Adversarial Examples Driven by Frequency. (99%)
Ti-Patch: Tiled Physical Adversarial Patch for no-reference video quality metrics. (83%)
Improving Weakly-Supervised Object Localization Using Adversarial Erasing and Pseudo Label. (1%)
Enhancing Code Vulnerability Detection via Vulnerability-Preserving Data Augmentation. (1%)
Consistency and Uncertainty: Identifying Unreliable Responses From Black-Box Vision-Language Models for Selective Visual Question Answering. (1%)
2024-04-14
Counteracting Concept Drift by Learning with Future Malware Predictions. (96%)
Watermark-embedded Adversarial Examples for Copyright Protection against Diffusion Models. (96%)
Adversarial Robustness Limits via Scaling-Law and Human-Alignment Studies. (76%)
FaceCat: Enhancing Face Recognition Security with a Unified Generative Model Framework. (22%)
2024-04-13
Stability and Generalization in Free Adversarial Training. (96%)
Proof-of-Learning with Incentive Security. (2%)
2024-04-12
PASA: Attack Agnostic Unsupervised Adversarial Detection using Prediction & Attribution Sensitivity Analysis. (99%)
Counterfactual Explanations for Face Forgery Detection via Adversarial Removal of Artifacts. (99%)
Struggle with Adversarial Defense? Try Diffusion. (99%)
Multimodal Attack Detection for Action Recognition Models. (83%)
A Survey of Neural Network Robustness Assessment in Image Recognition. (83%)
Practical Region-level Attack against Segment Anything Models. (81%)
FCert: Certifiably Robust Few-Shot Classification in the Era of Foundation Models. (69%)
Mitigating Cascading Effects in Large Adversarial Graph Environments. (2%)
On the Robustness of Language Guidance for Low-Level Vision Tasks: Findings from Depth Estimation. (1%)
Empowering Malware Detection Efficiency within Processing-in-Memory Architecture. (1%)
2024-04-11
Persistent Classification: A New Approach to Stability of Data and Adversarial Examples. (98%)
Eliminating Catastrophic Overfitting Via Abnormal Adversarial Examples Regularization. (98%)
Backdoor Contrastive Learning via Bi-level Trigger Optimization. (96%)
Adversarial Robustness of Distilled and Pruned Deep Learning-based Wireless Classifiers. (92%)
CodeFort: Robust Training for Code Generation Models. (33%)
AmpleGCG: Learning a Universal and Transferable Generative Model of Adversarial Suffixes for Jailbreaking Both Open and Closed LLMs. (12%)
LeapFrog: The Rowhammer Instruction Skip Attack. (8%)
Scaling (Down) CLIP: A Comprehensive Analysis of Data, Architecture, and Training Strategies. (1%)
2024-04-10
Logit Calibration and Feature Contrast for Robust Federated Learning on Non-IID Data. (99%)
Lost in Translation: Modern Neural Networks Still Struggle With Small Realistic Image Transformations. (82%)
Adversarial purification for no-reference image-quality metrics: applicability study and new methods. (26%)
Simpler becomes Harder: Do LLMs Exhibit a Coherent Behavior on Simplified Corpora? (2%)
TrajPRed: Trajectory Prediction with Region-based Relation Learning. (1%)
2024-04-09
Towards Building a Robust Toxicity Predictor. (99%)
On adversarial training and the 1 Nearest Neighbor classifier. (99%)
LRR: Language-Driven Resamplable Continuous Representation against Adversarial Tracking Attacks. (80%)
Towards Robust Domain Generation Algorithm Classification. (80%)
SafeGen: Mitigating Unsafe Content Generation in Text-to-Image Models. (41%)
Sandwich attack: Multi-language Mixture Adaptive Attack on LLMs. (31%)
Aggressive or Imperceptible, or Both: Network Pruning Assisted Hybrid Byzantines in Federated Learning. (26%)
How to Craft Backdoors with Unlabeled Data Alone? (1%)
2024-04-08
Certified PEFTSmoothing: Parameter-Efficient Fine-Tuning with Randomized Smoothing. (99%)
David and Goliath: An Empirical Evaluation of Attacks and Defenses for QNNs at the Deep Edge. (99%)
BruSLeAttack: A Query-Efficient Score-Based Black-Box Sparse Adversarial Attack. (99%)
Case Study: Neural Network Malware Detection Verification for Feature and Image Datasets. (98%)
Out-of-Distribution Data: An Acquaintance of Adversarial Examples -- A Survey. (98%)
Quantum Adversarial Learning for Kernel Methods. (75%)
Investigating the Impact of Quantization on Adversarial Robustness. (50%)
SphereHead: Stable 3D Full-head Synthesis with Spherical Tri-plane Representation. (1%)
2024-04-07
Semantic Stealth: Adversarial Text Attacks on NLP Using Several Methods. (99%)
Enabling Privacy-Preserving Cyber Threat Detection with Federated Learning. (15%)
How much reliable is ChatGPT's prediction on Information Extraction under Input Perturbations? (5%)
SemEval-2024 Task 2: Safe Biomedical Natural Language Inference for Clinical Trials. (1%)
2024-04-06
CANEDERLI: On The Impact of Adversarial Training and Transferability on CAN Intrusion Detection Systems. (86%)
Learning Minimal NAP Specifications for Neural Network Verification. (80%)
Data Poisoning Attacks on Off-Policy Policy Evaluation Methods. (67%)
Goal-guided Generative Prompt Injection Attack on Large Language Models. (67%)
Structured Gradient-based Interpretations via Norm-Regularized Adversarial Training. (61%)
Exploiting Sequence Number Leakage: TCP Hijacking in NAT-Enabled Wi-Fi Networks. (3%)
2024-04-05
Evaluating Adversarial Robustness: A Comparison Of FGSM, Carlini-Wagner Attacks, And The Role of Distillation as Defense Mechanism. (99%)
Reliable Feature Selection for Adversarially Robust Cyber-Attack Detection. (98%)
DiffuseMix: Label-Preserving Data Augmentation with Diffusion Models. (15%)
Compositional Estimation of Lipschitz Constants for Deep Neural Networks. (13%)
Precision Guided Approach to Mitigate Data Poisoning Attacks in Federated Learning. (12%)
2024-04-04
Meta Invariance Defense Towards Generalizable Robustness to Unknown Adversarial Attacks. (99%)
FACTUAL: A Novel Framework for Contrastive Learning Based Robust SAR Image Classification. (98%)
Learn What You Want to Unlearn: Unlearning Inversion Attacks against Machine Unlearning. (16%)
Knowledge Distillation-Based Model Extraction Attack using GAN-based Private Counterfactual Explanations. (8%)
Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks? (2%)
2024-04-03
Adversarial Attacks and Dimensionality in Text Classifiers. (99%)
Unsegment Anything by Simulating Deformation. (97%)
"Are Adversarial Phishing Webpages a Threat in Reality?" Understanding the Users' Perception of Adversarial Webpages. (81%)
JailBreakV-28K: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks. (75%)
Learn to Disguise: Avoid Refusal Responses in LLM's Defense via a Multi-agent Attacker-Disguiser Game. (11%)
A Unified Membership Inference Method for Visual Self-supervised Encoder via Part-aware Capability. (9%)
Steganographic Passport: An Owner and User Verifiable Credential for Deep Model IP Protection Without Retraining. (1%)
2024-04-02
Humanizing Machine-Generated Content: Evading AI-Text Detection through Adversarial Attack. (99%)
ADVREPAIR:Provable Repair of Adversarial Attack. (99%)
Jailbreaking Prompt Attack: A Controllable Adversarial Attack against Diffusion Models. (97%)
One Noise to Rule Them All: Multi-View Adversarial Attacks with Universal Perturbation. (92%)
Defense without Forgetting: Continual Adversarial Defense with Anisotropic & Isotropic Pseudo Replay. (88%)
Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks. (83%)
READ: Improving Relation Extraction from an ADversarial Perspective. (81%)
Two Heads are Better than One: Nested PoE for Robust Defense Against Multi-Backdoors. (64%)
Red-Teaming Segment Anything Model. (45%)
Towards Robust 3D Pose Transfer with Adversarial Learning. (31%)
Designing a Photonic Physically Unclonable Function Having Resilience to Machine Learning Attacks. (12%)
Exploring Backdoor Vulnerabilities of Chat Models. (2%)
CAPE: CAM as a Probabilistic Ensemble for Enhanced DNN Interpretation. (1%)
2024-04-01
The Double-Edged Sword of Input Perturbations to Robust Accurate Fairness. (99%)
Multi-granular Adversarial Attacks against Black-box Neural Ranking Models. (99%)
BadPart: Unified Black-box Adversarial Patch Attacks against Pixel-wise Regression Tasks. (93%)
Poisoning Decentralized Collaborative Recommender System and Its Countermeasures. (33%)
Can Biases in ImageNet Models Explain Generalization? (10%)
UFID: A Unified Framework for Input-level Backdoor Detection on Diffusion Models. (10%)
Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Models. (2%)
An incremental hybrid adaptive network-based IDS in Software Defined Networks to detect stealth attacks. (1%)
2024-03-31
PID Control-Based Self-Healing to Improve the Robustness of Large Language Models. (75%)
Machine Learning Robustness: A Primer. (62%)
2024-03-30
STBA: Towards Evaluating the Robustness of DNNs for Query-Limited Black-box Scenario. (99%)
Embodied Active Defense: Leveraging Recurrent Feedback to Counter Adversarial Patches. (98%)
Shortcuts Arising from Contrast: Effective and Covert Clean-Label Attacks in Prompt-Based Learning. (5%)
2024-03-29
On Inherent Adversarial Robustness of Active Vision Systems. (99%)
Benchmarking the Robustness of Temporal Action Detection Models Against Temporal Corruptions. (68%)
Deepfake Sentry: Harnessing Ensemble Intelligence for Resilient Detection and Generalisation. (8%)
The Impact of Prompts on Zero-Shot Detection of AI-Generated Text. (2%)
GDA: Generalized Diffusion for Robust Test-time Adaptation. (1%)
Efficient Data-Free Model Stealing with Label Diversity. (1%)
Cross-Lingual Transfer Robustness to Lower-Resource Languages on Adversarial Datasets. (1%)
2024-03-28
Towards Understanding Dual BN In Hybrid Adversarial Training. (82%)
Improving Adversarial Data Collection by Supporting Annotators: Lessons from GAHD, a German Hate Speech Dataset. (82%)
On the Robustness of LDP Protocols for Numerical Attributes under Data Poisoning Attacks. (41%)
MedBN: Robust Test-Time Adaptation against Malicious Test Samples. (10%)
Imperceptible Protection against Style Imitation from Diffusion Models. (2%)
A Backdoor Approach with Inverted Labels Using Dirty Label-Flipping Attacks. (1%)
2024-03-27
Uncertainty-Aware SAR ATR: Defending Against Adversarial Attacks via Bayesian Neural Networks. (99%)
CosalPure: Learning Concept from Group Images for Robust Co-Saliency Detection. (99%)
MMCert: Provable Defense against Adversarial Attacks to Multi-modal Models. (98%)
Bayesian Learned Models Can Detect Adversarial Malware For Free. (97%)
MisGUIDE : Defense Against Data-Free Deep Learning Model Extraction. (95%)
Towards Sustainable SecureML: Quantifying Carbon Footprint of Adversarial Machine Learning. (83%)
Deep Learning for Robust and Explainable Models in Computer Vision. (82%)
SemRoDe: Macro Adversarial Training to Learn Representations That are Robust to Word-Level Attacks. (81%)
JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models. (54%)
Vulnerability Detection with Code Language Models: How Far Are We? (26%)
Spikewhisper: Temporal Spike Backdoor Attacks on Federated Neuromorphic Learning over Low-power Devices. (15%)
Robustness and Visual Explanation for Black Box Image, Video, and ECG Signal Classification with Reinforcement Learning. (15%)
The Impact of Uniform Inputs on Activation Sparsity and Energy-Latency Attacks in Computer Vision. (11%)
Fact Checking Beyond Training Set. (1%)
BAM: Box Abstraction Monitors for Real-time OoD Detection in Object Detection. (1%)
2024-03-26
DataCook: Crafting Anti-Adversarial Examples for Healthcare Data Copyright Protection. (92%)
FaultGuard: A Generative Approach to Resilient Fault Prediction in Smart Electrical Grids. (78%)
Boosting Adversarial Training via Fisher-Rao Norm-based Regularization. (69%)
Optimization-based Prompt Injection Attack to LLM-as-a-Judge. (45%)
Targeted Visualization of the Backbone of Encoder LLMs. (9%)
Leak and Learn: An Attacker's Cookbook to Train Using Leaked Data from Federated Learning. (1%)
Exploring LLMs as a Source of Targeted Synthetic Textual Data to Minimize High Confidence Misclassifications. (1%)
2024-03-25
$\textit{LinkPrompt}$: Natural and Universal Adversarial Attacks on Prompt-based Language Models. (99%)
Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving. (98%)
The Anatomy of Adversarial Attacks: Concept-based XAI Dissection. (87%)
DeepKnowledge: Generalisation-Driven Deep Learning Testing. (82%)
Revealing Vulnerabilities of Neural Networks in Parameter Learning and Defense Against Explanation-Aware Backdoors. (70%)
LOTUS: Evasive and Resilient Backdoor Attacks through Sub-Partitioning. (69%)
Model-less Is the Best Model: Generating Pure Code Implementations to Replace On-Device DL Models. (1%)
2024-03-24
Subspace Defense: Discarding Adversarial Perturbations by Learning a Subspace for Clean Signals. (99%)
Ensemble Adversarial Defense via Integration of Multiple Dispersed Low Curvature Models. (98%)
Robust Diffusion Models for Adversarial Purification. (83%)
Unlearning Backdoor Threats: Enhancing Backdoor Defense in Multimodal Contrastive Learning via Local Token Unlearning. (5%)
Rumor Detection with a novel graph neural network approach. (4%)
Generating Potent Poisons and Backdoors from Scratch with Guided Diffusion. (2%)
A General and Efficient Federated Split Learning with Pre-trained Image Transformers for Heterogeneous Data. (1%)
2024-03-23
Towards Adversarial Robustness And Backdoor Mitigation in SSL. (76%)
Adversarial Defense Teacher for Cross-Domain Object Detection under Poor Visibility Conditions. (64%)
2024-03-22
Robust optimization for adversarial learning with finite sample complexity guarantees. (96%)
A Transfer Attack to Image Watermarks. (95%)
From Hardware Fingerprint to Access Token: Enhancing the Authentication on IoT Devices. (26%)
Clean-image Backdoor Attacks. (12%)
Forward Learning for Gradient-based Black-box Saliency Map Generation. (1%)
2024-03-21
Diffusion Attack: Leveraging Stable Diffusion for Naturalistic Image Attacking. (99%)
Few-Shot Adversarial Prompt Learning on Vision-Language Models. (98%)
Reversible Jump Attack to Textual Classifiers with Modification Reduction. (98%)
Improving Robustness to Model Inversion Attacks via Sparse Coding Architectures. (82%)
Adversary-Robust Graph-Based Learning of WSIs. (45%)
Safeguarding Medical Image Segmentation Datasets against Unauthorized Training via Contour- and Texture-Aware Perturbations. (4%)
2024-03-20
FMM-Attack: A Flow-based Multi-modal Adversarial Attack on Video-based LLMs. (97%)
DD-RobustBench: An Adversarial Robustness Benchmark for Dataset Distillation. (96%)
Capsule Neural Networks as Noise Stabilizer for Time Series Data. (93%)
Adversarial Attacks and Defenses in Automated Control Systems: A Comprehensive Benchmark. (70%)
Certified Human Trajectory Prediction. (61%)
Have You Poisoned My Data? Defending Neural Networks against Data Poisoning. (54%)
Mask-based Invisible Backdoor Attacks on Object Detection. (50%)
Defending Against Indirect Prompt Injection Attacks With Spotlighting. (31%)
Don't be a Fool: Pooling Strategies in Offensive Language Detection from User-Intended Adversarial Attacks. (11%)
BadEdit: Backdooring large language models by model editing. (1%)
Teacher-Student Training for Debiasing: General Permutation Debiasing for Large Language Models. (1%)
Threats, Attacks, and Defenses in Machine Unlearning: A Survey. (1%)
2024-03-19
As Firm As Their Foundations: Can open-sourced foundation models be used to create adversarial examples for downstream tasks? (99%)
Boosting Transferability in Vision-Language Attacks via Diversification along the Intersection Region of Adversarial Trajectory. (99%)
ADAPT to Robustify Prompt Tuning Vision Transformers. (98%)
Marlin: Knowledge-Driven Analysis of Provenance Graphs for Efficient and Robust Detection of Cyber Attacks. (75%)
Resilience in Online Federated Learning: Mitigating Model-Poisoning Attacks via Partial Sharing. (9%)
RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content. (8%)
Robust NAS under adversarial training: benchmark, theory, and beyond. (2%)
Discover and Mitigate Multiple Biased Subgroups in Image Classifiers. (1%)
2024-03-18
Diffusion Denoising as a Certified Defense against Clean-label Poisoning. (99%)
SSCAE -- Semantic, Syntactic, and Context-aware natural language Adversarial Examples generator. (99%)
LocalStyleFool: Regional Video Style Transfer Attack Using Segment Anything Model. (99%)
Invisible Backdoor Attack Through Singular Value Decomposition. (96%)
Problem space structural adversarial attacks for Network Intrusion Detection Systems based on Graph Neural Networks. (88%)
Impart: An Imperceptible and Effective Label-Specific Backdoor Attack. (83%)
SSAP: A Shape-Sensitive Adversarial Patch for Comprehensive Disruption of Monocular Depth Estimation in Autonomous Navigation Applications. (78%)
Electioneering the Network: Dynamic Multi-Step Adversarial Attacks for Community Canvassing. (61%)
Advancing Time Series Classification with Multimodal Language Modeling. (1%)
2024-03-17
Defense Against Adversarial Attacks on No-Reference Image Quality Models with Gradient Norm Regularization. (99%)
A Modified Word Saliency-Based Adversarial Attack on Text Classification Models. (99%)
Robust Overfitting Does Matter: Test-Time Adversarial Purification With FGSM. (99%)
Forging the Forger: An Attempt to Improve Authorship Verification via Data Augmentation. (76%)
RobustSentEmbed: Robust Sentence Embeddings Using Adversarial Self-Supervised Contrastive Learning. (50%)
COLEP: Certifiably Robust Learning-Reasoning Conformal Prediction via Probabilistic Circuits. (22%)
A Dual-Tier Adaptive One-Class Classification IDS for Emerging Cyberthreats. (9%)
Hierarchical Classification for Intrusion Detection System: Effective Design and Empirical Analysis. (2%)
CBR - Boosting Adaptive Classification By Retrieval of Encrypted Network Traffic with Out-of-distribution. (1%)
Pencil: Private and Extensible Collaborative Learning without the Non-Colluding Assumption. (1%)
2024-03-16
Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples. (98%)
Understanding Robustness of Visual State Space Models for Image Classification. (98%)
Improving Adversarial Transferability of Visual-Language Pre-training Models through Collaborative Multimodal Interaction. (92%)
Edge Private Graph Neural Networks with Singular Value Perturbation. (11%)
2024-03-15
Benchmarking Adversarial Robustness of Image Shadow Removal with Shadow-adaptive Attacks. (99%)
Towards Non-Adversarial Algorithmic Recourse. (99%)
Time-Frequency Jointed Imperceptible Adversarial Attack to Brainprint Recognition with Deep Learning Models. (99%)
Introducing Adaptive Continuous Adversarial Training (ACAT) to Enhance ML Robustness. (87%)
Revisiting Adversarial Training under Long-Tailed Distributions. (80%)
Towards Adversarially Robust Dataset Distillation by Curvature Regularization. (54%)
Interactive Trimming against Evasive Online Data Manipulation Attacks: A Game-Theoretic Approach. (50%)
Securing Federated Learning with Control-Flow Attestation: A Novel Framework for Enhanced Integrity and Resilience against Adversarial Attacks. (12%)
Benchmarking Zero-Shot Robustness of Multimodal Foundation Models: A Pilot Study. (11%)
Not Just Change the Labels, Learn the Features: Watermarking Deep Neural Networks with Multi-View Data. (8%)
Backdoor Secrets Unveiled: Identifying Backdoor Data with Optimized Scaled Prediction Consistency. (3%)
Robust Influence-based Training Methods for Noisy Brain MRI. (1%)
2024-03-14
An Image Is Worth 1000 Lies: Adversarial Transferability across Prompts on Vision-Language Models. (99%)
Counter-Samples: A Stateless Strategy to Neutralize Black Box Adversarial Attacks. (99%)
Adversarial Fine-tuning of Compressed Neural Networks for Joint Improvement of Robustness and Efficiency. (98%)
Soften to Defend: Towards Adversarial Robustness via Self-Guided Label Refinement. (83%)
Robust Subgraph Learning by Monitoring Early Training Representations. (80%)
LDPRecover: Recovering Frequencies from Poisoning Attacks against Local Differential Privacy. (76%)
AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shield Prompting. (74%)
Towards White Box Deep Learning. (15%)
Symbiotic Game and Foundation Models for Cyber Deception Operations in Strategic Cyber Warfare. (13%)
PreCurious: How Innocent Pre-Trained Language Models Turn into Privacy Traps. (12%)
AVIBench: Towards Evaluating the Robustness of Large Vision-Language Model on Adversarial Visual-Instructions. (2%)
Optimistic Verifiable Training by Controlling Hardware Nondeterminism. (1%)
Medical Unlearnable Examples: Securing Medical Data from Unauthorized Traning via Sparsity-Aware Local Masking. (1%)
ADEdgeDrop: Adversarial Edge Dropping for Robust Graph Neural Networks. (1%)
2024-03-13
Attack Deterministic Conditional Image Generative Models for Diverse and Controllable Generation. (92%)
Fast Inference of Removal-Based Node Influence. (54%)
Tastle: Distract Large Language Models for Automatic Jailbreak Attack. (31%)
Adaptive Hybrid Masking Strategy for Privacy-Preserving Face Recognition Against Model Inversion Attack. (8%)
RAF-GI: Towards Robust, Accurate and Fast-Convergent Gradient Inversion Attack in Federated Learning. (2%)
Verifix: Post-Training Correction to Improve Label Noise Robustness with Verified Samples. (1%)
2024-03-12
Versatile Defense Against Adversarial Attacks on Image Recognition. (99%)
Towards Model Extraction Attacks in GAN-Based Image Translation via Domain Shift Mitigation. (61%)
Backdoor Attack with Mode Mixture Latent Modification. (8%)
Towards a Framework for Deep Learning Certification in Safety-Critical Applications Using Inherently Safe Design and Run-Time Error Detection. (2%)
Duwak: Dual Watermarks in Large Language Models. (2%)
Visual Privacy Auditing with Diffusion Models. (1%)
2024-03-11
Intra-Section Code Cave Injection for Adversarial Evasion Attacks on Windows PE Malware File. (99%)
epsilon-Mesh Attack: A Surface-based Adversarial Point Cloud Attack for Facial Expression Recognition. (99%)
PeerAiD: Improving Adversarial Distillation from a Specialized Peer Tutor. (98%)
Dynamic Perturbation-Adaptive Adversarial Training on Medical Image Classification. (97%)
Disentangling Policy from Offline Task Representation Learning via Adversarial Data Augmentation. (96%)
PCLD: Point Cloud Layerwise Diffusion for Adversarial Purification. (86%)
Overcoming the Paradox of Certified Training with Gaussian Smoothing. (83%)
Real is not True: Backdoor Attacks Against Deepfake Detection. (78%)
Improving deep learning with prior knowledge and cognitive models: A survey on enhancing explainability, adversarial robustness and zero-shot learning. (61%)
Stealing Part of a Production Language Model. (38%)
AS-FIBA: Adaptive Selective Frequency-Injection for Backdoor Attack on Deep Face Restoration. (9%)
A novel interface for adversarial trivia question-writing. (3%)
Towards the Uncharted: Density-Descending Feature Perturbation for Semi-supervised Semantic Segmentation. (2%)
Learning with Noisy Foundation Models. (1%)
DNNShield: Embedding Identifiers for Deep Neural Network Ownership Verification. (1%)
2024-03-10
A Zero Trust Framework for Realization and Defense Against Generative AI Attacks in Power Grid. (22%)
2024-03-09
Hard-label based Small Query Black-box Adversarial Attack. (99%)
IOI: Invisible One-Iteration Adversarial Attack on No-Reference Image- and Video-Quality Metrics. (83%)
iBA: Backdoor Attack on 3D Point Cloud via Reconstructing Itself. (82%)
Attacking Transformers with Feature Diversity Adversarial Perturbation. (70%)
2024-03-08
Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds. (99%)
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. (99%)
Exploring the Adversarial Frontier: Quantifying Robustness via Adversarial Hypervolume. (98%)
Prepared for the Worst: A Learning-Based Adversarial Attack for Resilience Analysis of the ICP Algorithm. (93%)
Adversarial Sparse Teacher: Defense Against Distillation-Based Model Stealing Attacks Using Adversarial Examples. (93%)
EVD4UAV: An Altitude-Sensitive Benchmark to Evade Vehicle Detection in UAV. (81%)
The Impact of Quantization on the Robustness of Transformer-based Text Classifiers. (45%)
EdgeLeakage: Membership Information Leakage in Distributed Edge Intelligence Systems. (38%)
Speech Robust Bench: A Robustness Benchmark For Speech Recognition. (1%)
2024-03-07
Defending Against Unforeseen Failure Modes with Latent Adversarial Training. (83%)
Fooling Neural Networks for Motion Forecasting via Adversarial Attacks. (33%)
Automatic and Universal Prompt Injection Attacks against Large Language Models. (31%)
ObjectCompose: Evaluating Resilience of Vision-Based Models on Object-to-Background Compositional Changes. (31%)
Cell reprogramming design by transfer learning of functional transcriptional networks. (1%)
Towards Robustness Analysis of E-Commerce Ranking System. (1%)
2024-03-06
Adversarial Infrared Geometry: Using Geometry to Perform Adversarial Attack against Infrared Pedestrian Detectors. (99%)
Improving Adversarial Training using Vulnerability-Aware Perturbation Budget. (99%)
Effect of Ambient-Intrinsic Dimension Gap on Adversarial Vulnerability. (92%)
Belief-Enriched Pessimistic Q-Learning against Adversarial State Perturbations. (16%)
On the Effectiveness of Distillation in Mitigating Backdoors in Pre-trained Encoder. (2%)
Verified Training for Counterfactual Explanation Robustness under Data Shift. (2%)
2024-03-05
Towards Robust Federated Learning via Logits Calibration on Non-IID Data. (99%)
Mitigating Label Flipping Attacks in Malicious URL Detectors Using Ensemble Trees. (96%)
Minimum Topology Attacks for Graph Neural Networks. (83%)
Federated Learning Under Attack: Exposing Vulnerabilities through Data Poisoning Attacks in Computer Networks. (82%)
A general approach to enhance the survivability of backdoor attacks by decision path coupling. (68%)
Robust Federated Learning Mitigates Client-side Training Data Distribution Inference Attacks. (61%)
Uplift Modeling for Target User Attacks on Recommender Systems. (12%)
FLGuard: Byzantine-Robust Federated Learning via Ensemble of Contrastive Models. (11%)
InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents. (11%)
XAI-Based Detection of Adversarial Attacks on Deepfake Detectors. (8%)
2024-03-04
Robustness Bounds on the Successful Adversarial Examples: Theory and Practice. (99%)
One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models. (99%)
Improving the Robustness of Object Detection and Classification AI models against Adversarial Patch Attacks. (99%)
COMMIT: Certifying Robustness of Multi-Sensor Fusion Systems against Semantic Attacks. (96%)
Inf2Guard: An Information-Theoretic Framework for Learning Privacy-Preserving Representations against Inference Attacks. (26%)
BSDP: Brain-inspired Streaming Dual-level Perturbations for Online Open World Object Detection. (16%)
Mirage: Defense against CrossPath Attacks in Software Defined Networks. (3%)
Bayesian Uncertainty Estimation by Hamiltonian Monte Carlo: Applications to Cardiac MRI Segmentation. (1%)
2024-03-03
GuardT2I: Defending Text-to-Image Models from Adversarial Prompts. (10%)
2024-03-02
SAR-AE-SFP: SAR Imagery Adversarial Example in Real Physics domain with Target Scattering Feature Parameters. (99%)
Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy. (68%)
Breaking Down the Defenses: A Comparative Survey of Attacks on Large Language Models. (56%)
AutoDefense: Multi-Agent LLM Defense against Jailbreak Attacks. (31%)
Adversarial Testing for Visual Grounding via Image-Aware Property Reduction. (11%)
Query Recovery from Easy to Hard: Jigsaw Attack against SSE. (2%)
Accelerating Greedy Coordinate Gradient via Probe Sampling. (1%)
2024-03-01
Robust Deep Reinforcement Learning Through Adversarial Attacks and Training : A Survey. (91%)
Resilience of Entropy Model in Distributed Neural Networks. (67%)
Attacking Delay-based PUFs with Minimal Adversary Model. (45%)
2024-02-29
Unraveling Adversarial Examples against Speaker Identification -- Techniques for Attack Detection and Victim Model Classification. (99%)
How to Train your Antivirus: RL-based Hardening through the Problem-Space. (99%)
On Robustness and Generalization of ML-Based Congestion Predictors to Valid and Imperceptible Perturbations. (88%)
Pointing out the Shortcomings of Relation Extraction Models with Semantically Motivated Adversarials. (76%)
Assessing Visually-Continuous Corruption Robustness of Neural Networks Relative to Human Performance. (38%)
Verification of Neural Networks' Global Robustness. (38%)
SynGhost: Imperceptible and Universal Task-agnostic Backdoor Attack in Pre-trained Language Models. (16%)
Here's a Free Lunch: Sanitizing Backdoored Models with Model Merge. (2%)
Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes. (1%)
Unveiling Typographic Deceptions: Insights of the Typographic Vulnerability in Large Vision-Language Model. (1%)
2024-02-28
Enhancing the "Immunity" of Mixture-of-Experts Networks for Adversarial Defense. (99%)
MPAT: Building Robust Deep Neural Networks against Textual Adversarial Attacks. (99%)
Catastrophic Overfitting: A Potential Blessing in Disguise. (98%)
Living-off-The-Land Reverse-Shell Detection by Informed Data Augmentation. (76%)
A New Era in LLM Security: Exploring Security Concerns in Real-World LLM-based Systems. (64%)
Making Them Ask and Answer: Jailbreaking Large Language Models in Few Queries via Disguise and Reconstruction. (33%)
Out-of-Distribution Detection using Neural Activation Prior. (1%)
2024-02-27
Robustness-Congruent Adversarial Training for Secure Machine Learning Model Updates. (99%)
Extreme Miscalibration and the Illusion of Adversarial Robustness. (99%)
Black-box Adversarial Attacks Against Image Quality Assessment Models. (99%)
Enhancing Tracking Robustness with Auxiliary Adversarial Defense Networks. (99%)
LLM-Resistant Math Word Problem Generation via Adversarial Attacks. (87%)
Breaking the Black-Box: Confidence-Guided Model Inversion Attack for Distribution Shift. (83%)
Model X-ray:Detecting Backdoored Models via Decision Boundary. (67%)
Towards Fairness-Aware Adversarial Learning. (11%)
Time-Restricted Double-Spending Attack on PoW-based Blockchains. (1%)
2024-02-26
Improving the JPEG-resistance of Adversarial Attacks on Face Recognition by Interpolation Smoothing. (99%)
Improving behavior based authentication against adversarial attack using XAI. (99%)
Adversarial example soups: averaging multiple adversarial examples improves transferability without increasing additional generation time. (99%)
A Curious Case of Remarkable Resilience to Gradient Attacks via Fully Convolutional and Differentiable Front End with a Skip Connection. (98%)
Adversarial Perturbations of Physical Signals. (92%)
Unveiling Vulnerability of Self-Attention. (87%)
Edge Detectors Can Make Deep Convolutional Neural Networks More Robust. (83%)
Investigating Deep Watermark Security: An Adversarial Transferability Perspective. (64%)
Defending LLMs against Jailbreaking Attacks via Backtranslation. (64%)
Rainbow Teaming: Open-Ended Generation of Diverse Adversarial Prompts. (56%)
Pandora's White-Box: Precise Training Data Detection and Extraction in Large Language Models. (50%)
WIPI: A New Web Threat for LLM-Driven Web Agents. (8%)
RoCoIns: Enhancing Robustness of Large Language Models through Code-Style Instructions. (4%)
An Innovative Information Theory-based Approach to Tackle and Enhance The Transparency in Phishing Detection. (1%)
2024-02-25
From Noise to Clarity: Unraveling the Adversarial Suffix of Large Language Model Attacks via Translation of Text Embeddings. (98%)
An Adversarial Robustness Benchmark for Enterprise Network Intrusion Detection. (92%)
Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing. (76%)
Adversarial-Robust Transfer Learning for Medical Imaging via Domain Assimilation. (73%)
Evaluating Robustness of Generative Search Engine on Adversarial Factual Questions. (13%)
DrAttack: Prompt Decomposition and Reconstruction Makes Powerful LLM Jailbreakers. (2%)
State-of-the-Art Approaches to Enhancing Privacy Preservation of Machine Learning Datasets: A Survey. (1%)
m2mKD: Module-to-Module Knowledge Distillation for Modular Transformers. (1%)
2024-02-24
PRP: Propagating Universal Perturbations to Attack Large Language Model Guard-Rails. (87%)
LLMs Can Defend Themselves Against Jailbreaking in a Practical Manner: A Vision Paper. (86%)
RAUCA: A Novel Physical Adversarial Attack on Vehicle Detectors via Robust and Accurate Camouflage Generation. (82%)
Towards Robust Image Stitching: An Adaptive Resistance Learning against Compatible Attacks. (76%)
Optimal Zero-Shot Detector for Multi-Armed Attacks. (50%)
Sparse MeZO: Less Parameters for Better Performance in Zeroth-Order LLM Fine-Tuning. (1%)
2024-02-23
Distilling Adversarial Robustness Using Heterogeneous Teachers. (99%)
Fast Adversarial Attacks on Language Models In One GPU Minute. (98%)
A Robust Defense against Adversarial Attacks on Deep Learning-based Malware Detectors via (De)Randomized Smoothing. (98%)
ProTIP: Probabilistic Robustness Verification on Text-to-Image Diffusion Models against Stochastic Perturbation. (93%)
On the Duality Between Sharpness-Aware Minimization and Adversarial Training. (92%)
Low-Frequency Black-Box Backdoor Attack via Evolutionary Algorithm. (87%)
Deep Networks Always Grok and Here is Why. (76%)
BSPA: Exploring Black-box Stealthy Prompt Attacks against Image Generators. (67%)
Reinforcement Learning-Based Approaches for Enhancing Security and Resilience in Smart Control: A Survey on Attack and Defense Methods. (61%)
Break the Breakout: Reinventing LM Defense Against Jailbreak Attacks with Self-Refinement. (5%)
Prime+Retouch: When Cache is Locked and Leaked. (2%)
TREC: APT Tactic / Technique Recognition via Few-Shot Provenance Subgraph Learning. (1%)
2024-02-22
SoK: Analyzing Adversarial Examples: A Framework to Study Adversary Knowledge. (99%)
Rethinking Invariance Regularization in Adversarial Training to Improve Robustness-Accuracy Trade-off. (98%)
Stop Reasoning! When Multimodal LLMs with Chain-of-Thought Reasoning Meets Adversarial Images. (93%)
Noise-BERT: A Unified Perturbation-Robust Framework with Noise Alignment Pre-training for Noisy Slot Filling Task. (83%)
Mitigating Fine-tuning Jailbreak Attack with Backdoor Enhanced Alignment. (75%)
Getting Serious about Humor: Crafting Humor Datasets with Unfunny Large Language Models. (26%)
2024-02-21
AttackGNN: Red-Teaming GNNs in Hardware Security Using Reinforcement Learning. (99%)
A Simple and Yet Fairly Effective Defense for Graph Neural Networks. (98%)
Adversarial Purification and Fine-tuning for Robust UDC Image Restoration. (98%)
Is LLM-as-a-Judge Robust? Investigating Universal Adversarial Attacks on Zero-shot LLM Assessment. (83%)
Robustness of Deep Neural Networks for Micro-Doppler Radar Classification. (80%)
Flexible Physical Camouflage Generation Based on a Differential Approach. (38%)
VL-Trojan: Multimodal Instruction Backdoor Attacks against Autoregressive Visual Language Models. (10%)
Semantic Mirror Jailbreak: Genetic Algorithm Based Jailbreak Prompts Against Open-source LLMs. (8%)
Coercing LLMs to do and reveal (almost) anything. (4%)
T-Stitch: Accelerating Sampling in Pre-Trained Diffusion Models with Trajectory Stitching. (1%)
2024-02-20
QuanTest: Entanglement-Guided Testing of Quantum Neural Network Systems. (92%)
Defending Jailbreak Prompts via In-Context Adversarial Game. (76%)
Round Trip Translation Defence against Large Language Model Jailbreaking Attacks. (74%)
Investigating the Impact of Model Instability on Explanations and Uncertainty. (69%)
A Comprehensive Study of Jailbreak Attack versus Defense for Large Language Models. (68%)
Learning to Poison Large Language Models During Instruction Tuning. (13%)
Stealthy Adversarial Attacks on Stochastic Multi-Armed Bandits. (3%)
The Wolf Within: Covert Injection of Malice into MLLM Societies via an MLLM Operative. (1%)
RITFIS: Robust input testing framework for LLMs-based intelligent software. (1%)
2024-02-19
Query-Based Adversarial Prompt Generation. (99%)
Adversarial Feature Alignment: Balancing Robustness and Accuracy in Deep Learning via Adversarial Training. (99%)
AICAttack: Adversarial Image Captioning Attack with Attention-Based Optimization. (99%)
An Adversarial Approach to Evaluating the Robustness of Event Identification Models. (98%)
Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies. (97%)
Stealing the Invisible: Unveiling Pre-Trained CNN Models through Adversarial Examples and Timing Side-Channels. (92%)
Attacks on Node Attributes in Graph Neural Networks. (83%)
Indiscriminate Data Poisoning Attacks on Pre-trained Feature Extractors. (68%)
Self-Guided Robust Graph Structure Refinement. (67%)
Robust CLIP: Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models. (50%)
Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning. (15%)
Robustness and Exploration of Variational and Machine Learning Approaches to Inverse Problems: An Overview. (1%)
Amplifying Training Data Exposure through Fine-Tuning with Pseudo-Labeled Memberships. (1%)
Privacy-Preserving Low-Rank Adaptation for Latent Diffusion Models. (1%)
2024-02-18
A Curious Case of Searching for the Correlation between Training Data and Adversarial Robustness of Transformer Textual Models. (93%)
Evaluating Adversarial Robustness of Low dose CT Recovery. (92%)
Evaluating Efficacy of Model Stealing Attacks and Defenses on Quantum Neural Networks. (83%)
The Effectiveness of Random Forgetting for Robust Generalization. (75%)
Poisoned Forgery Face: Towards Backdoor Attacks on Face Forgery Detection. (26%)
Poisoning Federated Recommender Systems with Fake Users. (5%)
SPML: A DSL for Defending Language Models Against Prompt Attacks. (1%)
Teacher as a Lenient Expert: Teacher-Agnostic Data-Free Knowledge Distillation. (1%)
2024-02-17
Maintaining Adversarial Robustness in Continuous Learning. (75%)
Be Persistent: Towards a Unified Solution for Mitigating Shortcuts in Deep Learning. (22%)
Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents. (8%)
VoltSchemer: Use Voltage Noise to Manipulate Your Wireless Charger. (2%)
2024-02-16
DART: A Principled Approach to Adversarially Robust Unsupervised Domain Adaptation. (99%)
Theoretical Understanding of Learning from Adversarial Perturbations. (98%)
Assessing biomedical knowledge robustness in large language models by query-efficient sampling attacks. (98%)
VQAttack: Transferable Adversarial Attacks on Visual Question Answering via Pre-trained Models. (92%)
The AI Security Pyramid of Pain. (47%)
AIM: Automated Input Set Minimization for Metamorphic Security Testing. (2%)
Universal Prompt Optimizer for Safe Text-to-Image Generation. (1%)
ToBlend: Token-Level Blending With an Ensemble of LLMs to Attack AI-Generated Text Detection. (1%)
2024-02-15
Camouflage is all you need: Evaluating and Enhancing Language Model Robustness Against Camouflage Adversarial Attacks. (62%)
On the Safety Concerns of Deploying LLMs/VLMs in Robotics: Highlighting the Risks and Vulnerabilities. (31%)
Backdoor Attack against One-Class Sequential Anomaly Detection Models. (9%)
A Trembling House of Cards? Mapping Adversarial Attacks against Language Agents. (5%)
FedRDF: A Robust and Dynamic Aggregation Function against Poisoning Attacks in Federated Learning. (3%)
Quantum-Inspired Analysis of Neural Network Vulnerabilities: The Role of Conjugate Variables in System Attacks. (1%)
2024-02-14
Exploring the Adversarial Capabilities of Large Language Models. (98%)
PAL: Proxy-Guided Black-Box Attack on Large Language Models. (92%)
Only My Model On My Data: A Privacy Preserving Approach Protecting one Model and Deceiving Unauthorized Black-Box Models. (92%)
How Secure Are Large Language Models (LLMs) for Navigation in Urban Environments? (80%)
Review-Incorporated Model-Agnostic Profile Injection Attacks on Recommender Systems. (76%)
Attacking Large Language Models with Projected Gradient Descent. (67%)
Detecting Adversarial Spectrum Attacks via Distance to Decision Boundary Statistics. (47%)
SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding. (38%)
Reward Poisoning Attack Against Offline Reinforcement Learning. (12%)
Rapid Adoption, Hidden Risks: The Dual Impact of Large Language Model Customization. (9%)
Adversarial Nibbler: An Open Red-Teaming Method for Identifying Diverse Harms in Text-to-Image Generation. (3%)
Ten Words Only Still Help: Improving Black-Box AI-Generated Text Detection via Proxy-Guided Efficient Re-Sampling. (2%)
Leveraging the Context through Multi-Round Interactions for Jailbreaking Attacks. (1%)
Towards Robust Model-Based Reinforcement Learning Against Adversarial Corruption. (1%)
Immediate generalisation in humans but a generalisation lag in deep neural networks$\unicode{x2014}$evidence for representational divergence? (1%)
Play Guessing Game with LLM: Indirect Jailbreak Attack with Implicit Clues. (1%)
2024-02-13
Faster Repeated Evasion Attacks in Tree Ensembles. (96%)
Generating Universal Adversarial Perturbations for Quantum Classifiers. (93%)
Enhancing Robustness of Indoor Robotic Navigation with Free-Space Segmentation Models Against Adversarial Attacks. (83%)
Data Reconstruction Attacks and Defenses: A Systematic Evaluation. (76%)
COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability. (62%)
Test-Time Backdoor Attacks on Multimodal Large Language Models. (56%)
Adversarially Robust Feature Learning for Breast Cancer Diagnosis. (33%)
Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast. (31%)
Adaptive Hierarchical Certification for Segmentation using Randomized Smoothing. (10%)
Feature Attribution with Necessity and Sufficiency via Dual-stage Perturbation Test for Causal Explanation. (1%)
2024-02-12
Understanding Deep Learning defenses Against Adversarial Examples Through Visualizations for Dynamic Risk Assessment. (99%)
Topological safeguard for evasion attack interpreting the neural networks' behavior. (89%)
PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models. (83%)
Privacy-Preserving Gaze Data Streaming in Immersive Interactive Virtual Reality: Robustness and User Experience. (33%)
OrderBkd: Textual backdoor attack through repositioning. (13%)
Tighter Bounds on the Information Bottleneck with Application to Deep Learning. (10%)
Multi-Attribute Vision Transformers are Efficient and Robust Learners. (9%)
Customizable Perturbation Synthesis for Robust SLAM Benchmarking. (9%)
THE COLOSSEUM: A Benchmark for Evaluating Generalization for Robotic Manipulation. (5%)
Accelerated Smoothing: A Scalable Approach to Randomized Smoothing. (3%)
Game of Trojans: Adaptive Adversaries Against Output-based Trojaned-Model Detectors. (3%)
Local Centrality Minimization with Quality Guarantees. (1%)
NeuralSentinel: Safeguarding Neural Network Reliability and Trustworthiness. (1%)
Do Membership Inference Attacks Work on Large Language Models? (1%)
Pixel Sentence Representation Learning. (1%)
2024-02-11
A Random Ensemble of Encrypted Vision Transformers for Adversarially Robust Defense. (99%)
Accuracy of TextFooler black box adversarial attacks on 01 loss sign activation neural network ensemble. (98%)
2024-02-10
Whispers in the Machine: Confidentiality in LLM-integrated Systems. (26%)
Architectural Neural Backdoors from First Principles. (26%)
2024-02-09
Anomaly Unveiled: Securing Image Classification against Adversarial Patch Attacks. (98%)
Fight Back Against Jailbreaking via Prompt Adversarial Tuning. (95%)
RAMP: Boosting Adversarial Robustness Against Multiple $l_p$ Perturbations. (84%)
System-level Analysis of Adversarial Attacks and Defenses on Intelligence in O-RAN based Cellular Networks. (82%)
The SkipSponge Attack: Sponge Weight Poisoning of Deep Neural Networks. (70%)
Corruption Robust Offline Reinforcement Learning with Human Feedback. (67%)
Quantifying and Enhancing Multi-modal Robustness with Modality Preference. (56%)
StruQ: Defending Against Prompt Injection with Structured Queries. (45%)
Evaluating Membership Inference Attacks and Defenses in Federated Learning. (4%)
Blockchain Bribing Attacks and the Efficacy of Counterincentives. (1%)
For Better or For Worse? Learning Minimum Variance Features With Label Augmentation. (1%)
2024-02-08
Comprehensive Assessment of Jailbreak Attacks Against LLMs. (99%)
Investigating White-Box Attacks for On-Device Models. (93%)
TETRIS: Towards Exploring the Robustness of Interactive Segmentation. (81%)
Linearizing Models for Efficient yet Robust Private Inference. (68%)
A High Dimensional Statistical Model for Adversarial Training: Geometry and Trade-Offs. (26%)
Is Adversarial Training with Compressed Datasets Effective? (10%)
Reinforcement Learning as a Catalyst for Robust and Fair Federated Learning: Deciphering the Dynamics of Client Contributions. (9%)
2024-02-07
Adversarial Robustness Through Artifact Design. (99%)
Breaking Free: How to Hack Safety Guardrails in Black-Box Diffusion Models! (98%)
Analyzing Adversarial Inputs in Deep Reinforcement Learning. (96%)
Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications. (1%)
2024-02-06
Boosting Adversarial Transferability across Model Genus by Deformation-Constrained Warping. (98%)
Adversarially Robust Deepfake Detection via Adversarial Feature Similarity Learning. (98%)
SUB-PLAY: Adversarial Policies against Partially Observed Multi-Agent Reinforcement Learning Systems. (76%)
PAC-Bayesian Adversarially Robust Generalization Bounds for Graph Neural Network. (75%)
Enhance DNN Adversarial Robustness and Efficiency via Injecting Noise to Non-Essential Neurons. (74%)
BotSSCL: Social Bot Detection with Self-Supervised Contrastive Learning. (64%)
Privacy Leakage on DNNs: A Survey of Model Inversion Attacks and Defenses. (26%)
Studying Vulnerable Code Entities in R. (10%)
DeMarking: A Defense for Network Flow Watermarking in Real-Time. (10%)
HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal. (2%)
2024-02-05
A Generative Approach to Surrogate-based Black-box Attacks. (99%)
Transcending Adversarial Perturbations: Manifold-Aided Adversarial Examples with Legitimate Semantics. (99%)
Arabic Synonym BERT-based Adversarial Examples for Text Classification. (99%)
Generalization Properties of Adversarial Training for $\ell_0$-Bounded Adversarial Attacks. (92%)
FoolSDEdit: Deceptively Steering Your Edits Towards Targeted Attribute-aware Distribution. (89%)
Time-Distributed Backdoor Attacks on Federated Spiking Learning. (83%)
Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models. (83%)
Partially Recentralization Softmax Loss for Vision-Language Models Robustness. (81%)
Organic or Diffused: Can We Distinguish Human Art from AI-generated Images? (31%)
DisDet: Exploring Detectability of Backdoor Attack on Diffusion Models. (12%)
FINEST: Stabilizing Recommendations by Rank-Preserving Fine-Tuning. (1%)
Exploring mechanisms of Neural Robustness: probing the bridge between geometry and spectrum. (1%)
2024-02-04
PROSAC: Provably Safe Certification for Machine Learning Models under Adversarial Attacks. (99%)
Adversarial Text Purification: A Large Language Model Approach for Defense. (99%)
DeSparsify: Adversarial Attack Against Token Sparsification Mechanisms in Vision Transformers. (99%)
Exploiting Class Probabilities for Black-box Sentence-level Attacks. (75%)
Evading Deep Learning-Based Malware Detectors via Obfuscation: A Deep Reinforcement Learning Approach. (41%)
Adversarial Data Augmentation for Robust Speaker Verification. (1%)
2024-02-03
Seeing is not always believing: The Space of Harmless Perturbations. (99%)
Evaluating the Robustness of Off-Road Autonomous Driving Segmentation against Adversarial Attacks: A Dataset-Centric analysis. (96%)
Your Diffusion Model is Secretly a Certifiably Robust Classifier. (91%)
MixedNUTS: Training-Free Accuracy-Robustness Balance via Nonlinearly Mixed Classifiers. (76%)
Analyzing Sentiment Polarity Reduction in News Presentation through Contextual Perturbation and Large Language Models. (68%)
Universal Post-Training Reverse-Engineering Defense Against Backdoors in Deep Neural Networks. (31%)
Towards Optimal Adversarial Robust Q-learning with Bellman Infinity-error. (10%)
Invisible Finger: Practical Electromagnetic Interference Attack on Touchscreen-based Electronic Devices. (9%)
Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models. (5%)
Data Poisoning for In-context Learning. (5%)
2024-02-02
HQA-Attack: Toward High Quality Black-Box Hard-Label Adversarial Attack on Text. (99%)
$\sigma$-zero: Gradient-based Optimization of $\ell_0$-norm Adversarial Examples. (99%)
STAA-Net: A Sparse and Transferable Adversarial Attack for Speech Emotion Recognition. (99%)
Delving into Decision-based Black-box Attacks on Semantic Segmentation. (93%)
SignSGD with Federated Defense: Harnessing Adversarial Attacks through Gradient Sign Decoding. (92%)
Unlearnable Examples For Time Series. (86%)
Preference Poisoning Attacks on Reward Model Learning. (83%)
Privacy-Preserving Distributed Learning for Residential Short-Term Load Forecasting. (3%)
S2malloc: Statistically Secure Allocator for Use-After-Free Protection And More. (3%)
Cheating Suffix: Targeted Attack to Text-To-Image Diffusion Models with Multi-Modal Priors. (2%)
Fundamental Challenges in Cybersecurity and a Philosophy of Vulnerability-Guided Hardening. (1%)
What Will My Model Forget? Forecasting Forgotten Examples in Language Model Refinement. (1%)
2024-02-01
Benchmarking Transferable Adversarial Attacks. (98%)
Hidding the Ghostwriters: An Adversarial Evaluation of AI-Generated Student Essay Detection. (70%)
Double-Dip: Thwarting Label-Only Membership Inference Attacks with Transfer Learning and Randomization. (64%)
Vision-LLMs Can Fool Themselves with Self-Generated Typographic Attacks. (45%)
Approximating Optimal Morphing Attacks using Template Inversion. (9%)
Trustworthy Distributed AI Systems: Robustness, Privacy, and Governance. (8%)
Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attack. (1%)
algoXSSF: Detection and analysis of cross-site request forgery (XSRF) and cross-site scripting (XSS) attacks via Machine learning algorithms. (1%)
2024-01-31
Adversarial Quantum Machine Learning: An Information-Theoretic Generalization Analysis. (95%)
Invariance-powered Trustworthy Defense via Remove Then Restore. (70%)
BrainLeaks: On the Privacy-Preserving Properties of Neuromorphic Architectures against Model Inversion Attacks. (13%)
LoRec: Large Language Model for Robust Sequential Recommendation against Poisoning Attacks. (9%)
Logit Poisoning Attack in Distillation-based Federated Learning and its Countermeasures. (4%)
Manipulating Predictions over Discrete Inputs in Machine Teaching. (1%)
Ambush from All Sides: Understanding Security Threats in Open-Source Software CI/CD Pipelines. (1%)
2024-01-30
Single Word Change is All You Need: Designing Attacks and Defenses for Text Classifiers. (99%)
Robust Prompt Optimization for Defending Language Models Against Jailbreaking Attacks. (98%)
Towards Assessing the Synthetic-to-Measured Adversarial Vulnerability of SAR ATR. (98%)
AdvGPS: Adversarial GPS for Multi-Agent Perception Attack. (95%)
Game-Theoretic Unlearnable Example Generator. (92%)
Camouflage Adversarial Attacks on Multiple Agent Systems. (87%)
Weak-to-Strong Jailbreaking on Large Language Models. (76%)
A Proactive and Dual Prevention Mechanism against Illegal Song Covers empowered by Singing Voice Conversion. (75%)
Improving QA Model Performance with Cartographic Inoculation. (26%)
Towards Visual Syntactical Understanding. (4%)
Provably Robust Multi-bit Watermarking for AI-generated Text via Error Correction Code. (2%)
2024-01-29
LESSON: Multi-Label Adversarial False Data Injection Attack for Deep Learning Locational Detection. (99%)
Adversarial Training on Purification (AToP): Advancing Both Robustness and Generalization. (92%)
Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks. (68%)
GPS: Graph Contrastive Learning via Multi-scale Augmented Views from Adversarial Pooling. (5%)
Security and Privacy Challenges of Large Language Models: A Survey. (1%)
2024-01-28
Addressing Noise and Efficiency Issues in Graph-Based Machine Learning Models From the Perspective of Adversarial Attack. (83%)
Transparency Attacks: How Imperceptible Image Layers Can Fool AI Perception. (75%)
Model Supply Chain Poisoning: Backdooring Pre-trained Models via Embedding Indistinguishability. (26%)
2024-01-27
L-AutoDA: Leveraging Large Language Models for Automated Decision-based Adversarial Attacks. (98%)
2024-01-26
Set-Based Training for Neural Network Verification. (99%)
Mitigating Feature Gap for Adversarial Robustness by Feature Disentanglement. (91%)
Multi-Trigger Backdoor Attacks: More Triggers, More Threats. (82%)
Adversarial Attacks and Defenses in 6G Network-Assisted IoT Systems. (81%)
Conserve-Update-Revise to Cure Generalization and Robustness Trade-off in Adversarial Training. (62%)
Asymptotic Behavior of Adversarial Training Estimator under $\ell_\infty$-Perturbation. (22%)
Better Representations via Adversarial Training in Pre-Training: A Theoretical Perspective. (22%)
MEA-Defender: A Robust Watermark against Model Extraction Attack. (13%)
BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learning. (2%)
2024-01-25
Sparse and Transferable Universal Singular Vectors Attack. (99%)
Friendly Attacks to Improve Channel Coding Reliability. (54%)
Semantic Sensitivities and Inconsistent Predictions: Measuring the Fragility of NLI Models. (16%)
The Risk of Federated Learning to Skew Fine-Tuning Features and Underperform Out-of-Distribution Robustness. (2%)
Novel Quadratic Constraints for Extending LipSDP beyond Slope-Restricted Activations. (1%)
Physical Trajectory Inference Attack and Defense in Decentralized POI Recommendation. (1%)
2024-01-24
A Training Rate and Survival Heuristic for Inference and Robustness Evaluation (TRASHFIRE). (92%)
Can overfitted deep neural networks in adversarial training generalize? -- An approximation viewpoint. (86%)
WPDA: Frequency-based Backdoor Attack with Wavelet Packet Decomposition. (76%)
Exploring Adversarial Threat Models in Cyber Physical Battery Systems. (76%)
Fluent dreaming for language models. (64%)
2024-01-23
Boosting the Transferability of Adversarial Examples via Local Mixup and Adaptive Step Size. (99%)
Securing Recommender System via Cooperative Training. (80%)
Compositional Generative Inverse Design. (56%)
AdCorDA: Classifier Refinement via Adversarial Correction and Domain Adaptation. (33%)
ToDA: Target-oriented Diffusion Attacker against Recommendation System. (13%)
DAFA: Distance-Aware Fair Adversarial Training. (2%)
The twin peaks of learning neural networks. (2%)
2024-01-22
Fast Adversarial Training against Textual Adversarial Attacks. (99%)
A Training-Free Defense Framework for Robust Learned Image Compression. (74%)
Adversarial speech for voice privacy protection from Personalized Speech generation. (73%)
NEUROSEC: FPGA-Based Neuromorphic Audio Security. (13%)
Unraveling Attacks in Machine Learning-based IoT Ecosystems: A Survey and the Open Libraries Behind Them. (13%)
Robustness to distribution shifts of compressed networks for edge devices. (8%)
Text Embedding Inversion Security for Multilingual Language Models. (2%)
Out-of-Distribution Detection & Applications With Ablated Learned Temperature Energy. (1%)
2024-01-21
How Robust Are Energy-Based Models Trained With Equilibrium Propagation? (99%)
Analyzing the Quality Attributes of AI Vision Models in Open Repositories Under Adversarial Attacks. (56%)
Adversarial Augmentation Training Makes Action Recognition Models More Robust to Realistic Video Distribution Shifts. (11%)
Efficient local linearity regularization to overcome catastrophic overfitting. (8%)
2024-01-20
Susceptibility of Adversarial Attack on Medical Image Segmentation Models. (99%)
Finding a Needle in the Adversarial Haystack: A Targeted Paraphrasing Approach For Uncovering Edge Cases with Minimal Distribution Distortion. (96%)
CARE: Ensemble Adversarial Robustness Evaluation Against Adaptive Attackers for Security Applications. (80%)
Inducing High Energy-Latency of Large Vision-Language Models with Verbose Images. (33%)
2024-01-19
PuriDefense: Randomized Local Implicit Adversarial Purification for Defending Black-box Query-based Attacks. (99%)
Explainable and Transferable Adversarial Attack for ML-Based Network Intrusion Detectors. (99%)
The Surprising Harmfulness of Benign Overfitting for Adversarial Robustness. (98%)
FIMBA: Evaluating the Robustness of AI in Genomics via Feature Importance Adversarial Attacks. (56%)
Adversarially Robust Signed Graph Contrastive Learning from Balance Augmentation. (10%)
BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models. (3%)
Image Safeguarding: Reasoning with Conditional Vision Language Model and Obfuscating Unsafe Content Counterfactually. (1%)
2024-01-18
HGAttack: Transferable Heterogeneous Graph Adversarial Attack. (99%)
Hijacking Attacks against Neural Networks by Analyzing Training Data. (99%)
Adapters Mixup: Mixing Parameter-Efficient Adapters to Enhance the Adversarial Robustness of Fine-tuned Pre-trained Text Classifiers. (99%)
Hacking Predictors Means Hacking Cars: Using Sensitivity Analysis to Identify Trajectory Prediction Vulnerabilities for Autonomous Driving Security. (92%)
Differentially Private and Adversarially Robust Machine Learning: An Empirical Evaluation. (80%)
Investigating Training Strategies and Model Robustness of Low-Rank Adaptation for Language Modeling in Speech Recognition. (15%)
Power in Numbers: Robust reading comprehension by finetuning with four adversarial sentences per example. (13%)
Cross-Modality Perturbation Synergy Attack for Person Re-identification. (3%)
Vulnerabilities of Foundation Model Integrated Federated Learning Under Adversarial Threats. (2%)
Large Language Models are Efficient Learners of Noise-Robust Speech Recognition. (1%)
2024-01-17
Towards Scalable and Robust Model Versioning. (93%)
Artwork Protection Against Neural Style Transfer Using Locally Adaptive Adversarial Color Attack. (93%)
MITS-GAN: Safeguarding Medical Imaging from Tampering with Generative Adversarial Networks. (26%)
A GAN-based data poisoning framework against anomaly detection in vertical federated learning. (3%)
An Optimal Transport Approach for Computing Adversarial Training Lower Bounds in Multiclass Classification. (3%)
Attack and Reset for Unlearning: Exploiting Adversarial Noise toward Machine Unlearning through Parameter Re-initialization. (1%)
Caught in the Quicksand of Reasoning, Far from AGI Summit: Evaluating LLMs' Mathematical and Coding Competency through Ontology-guided Interventions. (1%)
2024-01-16
Revealing Vulnerabilities in Stable Diffusion via Targeted Attacks. (99%)
Bag of Tricks to Boost Adversarial Transferability. (99%)
A Generative Adversarial Attack for Multilingual Text Classifiers. (99%)
PPR: Enhancing Dodging Attacks while Maintaining Impersonation Attacks on Face Recognition Systems. (99%)
Robust Localization of Key Fob Using Channel Impulse Response of Ultra Wide Band Sensors for Keyless Entry Systems. (92%)
The Effect of Intrinsic Dataset Properties on Generalization: Unraveling Learning Differences Between Natural and Medical Images. (87%)
RandOhm: Mitigating Impedance Side-channel Attacks using Randomized Circuit Configurations. (9%)
Towards Efficient and Certified Recovery from Poisoning Attacks in Federated Learning. (8%)
IPR-NeRF: Ownership Verification meets Neural Radiance Field. (3%)
IoTWarden: A Deep Reinforcement Learning Based Real-time Defense System to Mitigate Trigger-action IoT Attacks. (1%)
2024-01-15
Robustness Against Adversarial Attacks via Learning Confined Adversarial Polytopes. (99%)
Authorship Obfuscation in Multilingual Machine-Generated Text Detection. (13%)
2024-01-14
LookAhead: Preventing DeFi Attacks via Unveiling Adversarial Contracts. (80%)
Crafter: Facial Feature Crafting against Inversion-based Identity Theft on Deep Models. (70%)
2024-01-13
Exploring Adversarial Attacks against Latent Diffusion Model from the Perspective of Adversarial Transferability. (99%)
Left-right Discrepancy for Adversarial Attack on Stereo Networks. (98%)
2024-01-12
Adversarial Examples are Misaligned in Diffusion Model Manifolds. (98%)
How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs. (2%)
Enhancing Consistency and Mitigating Bias: A Data Replay Approach for Incremental Learning. (1%)
An Analytical Framework for Modeling and Synthesizing Malicious Attacks on ACC Vehicles. (1%)
Intention Analysis Makes LLMs A Good Jailbreak Defender. (1%)
2024-01-11
GE-AdvGAN: Improving the transferability of adversarial samples by gradient editing-based adversarial generative model. (99%)
Universal Vulnerabilities in Large Language Models: In-context Learning Backdoor Attacks. (61%)
Open the Pandora's Box of LLMs: Jailbreaking LLMs through Representation Engineering. (22%)
Can We Trust the Unlabeled Target Data? Towards Backdoor Attack and Defense on Model Adaptation. (8%)
Manipulating Feature Visualizations with Gradient Slingshots. (3%)
Combating Adversarial Attacks with Multi-Agent Debate. (3%)
2024-01-10
Exploring Vulnerabilities of No-Reference Image Quality Assessment Models: A Query-Based Black-Box Method. (83%)
TrustLLM: Trustworthiness in Large Language Models. (75%)
SENet: Visual Detection of Online Social Engineering Attack Campaigns. (4%)
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training. (2%)
CoLafier: Collaborative Noisy Label Purifier With Local Intrinsic Dimensionality Guidance. (1%)
Brave: Byzantine-Resilient and Privacy-Preserving Peer-to-Peer Federated Learning. (1%)
FBSDetector: Fake Base Station and Multi Step Attack Detection in Cellular Networks using Machine Learning. (1%)
2024-01-09
Revisiting Adversarial Training at Scale. (26%)
SoK: Facial Deepfake Detectors. (11%)
Advancing Ante-Hoc Explainable Models through Generative Adversarial Networks. (3%)
2024-01-08
Pre-trained Model Guided Fine-Tuning for Zero-Shot Adversarial Robustness. (99%)
Robustness Assessment of a Runway Object Classifier for Safe Aircraft Taxiing. (54%)
Coupling Graph Neural Networks with Fractional Order Continuous Dynamics: A Robustness Study. (45%)
Logits Poisoning Attack in Federated Distillation. (12%)
Attack-Resilient Image Watermarking Using Stable Diffusion. (3%)
Dense Hopfield Networks in the Teacher-Student Setting. (1%)
2024-01-07
Invisible Reflections: Leveraging Infrared Laser Reflections to Target Traffic Sign Perception. (87%)
Data-Driven Subsampling in the Presence of an Adversarial Actor. (86%)
ROIC-DM: Robust Text Inference and Classification via Diffusion Model. (33%)
2024-01-06
Data-Dependent Stability Analysis of Adversarial Training. (98%)
End-to-End Anti-Backdoor Learning on Images and Time Series. (61%)
2024-01-05
Transferable Learned Image Compression-Resistant Adversarial Perturbations. (99%)
Enhancing targeted transferability via feature space fine-tuning. (98%)
Calibration Attack: A Framework For Adversarial Attacks Targeting Calibration. (76%)
A backdoor attack against link prediction tasks with graph neural networks. (38%)
TEN-GUARD: Tensor Decomposition for Backdoor Attack Detection in Deep Neural Networks. (1%)
MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance. (1%)
2024-01-04
Vulnerabilities Unveiled: Adversarially Attacking a Multimodal Vision Langauge Model for Pathology Imaging. (99%)
A Random Ensemble of Encrypted models for Enhancing Robustness against Adversarial Examples. (99%)
AdvSQLi: Generating Adversarial SQL Injections against Real-world WAF-as-a-service. (95%)
Evasive Hardware Trojan through Adversarial Power Trace. (92%)
Object-oriented backdoor attack against image captioning. (76%)
DEM: A Method for Certifying Deep Neural Network Classifier Outputs in Aerospace. (2%)
Secure Control of Connected and Automated Vehicles Using Trust-Aware Robust Event-Triggered Control Barrier Functions. (2%)
A Survey Analyzing Generalization in Deep Reinforcement Learning. (1%)
2024-01-03
Towards Robust Semantic Segmentation against Patch-based Attack via Attention Refinement. (92%)
Spy-Watermark: Robust Invisible Watermarking for Backdoor Attack. (62%)
FullLoRA-AT: Efficiently Boosting the Robustness of Pretrained Vision Transformers. (33%)
Integrated Cyber-Physical Resiliency for Power Grids under IoT-Enabled Dynamic Botnet Attacks. (22%)
Enhancing Generalization of Invisible Facial Privacy Cloak via Gradient Accumulation. (1%)
2024-01-02
JMA: a General Algorithm to Craft Nearly Optimal Targeted Adversarial Example. (99%)
Dual Teacher Knowledge Distillation with Domain Alignment for Face Anti-spoofing. (92%)
SpecFormer: Guarding Vision Transformer Robustness via Maximum Singular Value Penalization. (75%)
Unveiling the Stealthy Threat: Analyzing Slow Drift GPS Spoofing Attacks for Autonomous Vehicles in Urban Environments and Enabling the Resilience. (10%)
Imperio: Language-Guided Backdoor Attacks for Arbitrary Model Control. (4%)
Will 6G be Semantic Communications? Opportunities and Challenges from Task Oriented and Secure Communications to Integrated Sensing. (2%)
2024-01-01
Safety and Performance, Why Not Both? Bi-Objective Optimized Model Compression against Heterogeneous Attacks Toward AI Software Deployment. (12%)
Detection and Defense Against Prominent Attacks on Preconditioned LLM-Integrated Virtual Assistants. (8%)
A Novel Evaluation Framework for Assessing Resilience Against Prompt Injection Attacks in Large Language Models. (2%)
2023-12-31
AR-GAN: Generative Adversarial Network-Based Defense Method Against Adversarial Attacks on the Traffic Sign Classification System of Autonomous Vehicles. (99%)
Does Few-shot Learning Suffer from Backdoor Attacks? (98%)
Is It Possible to Backdoor Face Forgery Detection with Natural Triggers? (68%)
2023-12-30
Explainability-Driven Leaf Disease Classification using Adversarial Training and Knowledge Distillation. (84%)
CamPro: Camera-based Anti-Facial Recognition. (81%)
TPatch: A Triggered Physical Adversarial Patch. (76%)
A clean-label graph backdoor attack method in node classification task. (9%)
2023-12-29
Jatmo: Prompt Injection Defense by Task-Specific Finetuning. (54%)
SSL-OTA: Unveiling Backdoor Threats in Self-Supervised Learning for Object Detection. (11%)
Towards Faithful Explanations for Text Classification with Robustness Improvement and Explanation Guided Training. (9%)
2023-12-28
Adversarial Attacks on Image Classification Models: Analysis and Defense. (99%)
BlackboxBench: A Comprehensive Benchmark of Black-box Adversarial Attacks. (99%)
Attack Tree Analysis for Adversarial Evasion Attacks. (99%)
Can you See me? On the Visibility of NOPs against Android Malware Detectors. (98%)
MVPatch: More Vivid Patch for Adversarial Camouflaged Attacks on Object Detectors in the Physical World. (98%)
Explainability-Based Adversarial Attack on Graphs Through Edge Perturbation. (92%)
DOEPatch: Dynamically Optimized Ensemble Model for Adversarial Patches Generation. (83%)
Securing NextG Systems against Poisoning Attacks on Federated Learning: A Game-Theoretic Solution. (64%)
Timeliness: A New Design Metric and a New Attack Surface. (1%)
2023-12-27
Adversarial Attacks on LoRa Device Identification and Rogue Signal Detection with Deep Learning. (98%)
Domain Generalization with Vital Phase Augmentation. (3%)
2023-12-26
From text to multimodal: a survey of adversarial example generation in question answering systems. (92%)
Natural Adversarial Patch Generation Method Based on Latent Diffusion Model. (76%)
Robust Survival Analysis with Adversarial Regularization. (61%)
Universal Pyramid Adversarial Training for Improved ViT Performance. (5%)
2023-12-25
GanFinger: GAN-Based Fingerprint Generation for Deep Neural Network Ownership Verification. (96%)
Adversarial Item Promotion on Visually-Aware Recommender Systems by Guided Diffusion. (84%)
Punctuation Matters! Stealthy Backdoor Attack for Language Models. (11%)
2023-12-23
Adversarial Data Poisoning for Fake News Detection: How to Make a Model Misclassify a Target News without Modifying It. (10%)
Pre-trained Trojan Attacks for Visual Recognition. (1%)
TVE: Learning Meta-attribution for Transferable Vision Explainer. (1%)
2023-12-22
MEAOD: Model Extraction Attack against Object Detectors. (83%)
Asymmetric Bias in Text-to-Image Generation with Adversarial Attacks. (82%)
Understanding the Regularity of Self-Attention with Optimal Transport. (31%)
Attacking Byzantine Robust Aggregation in High Dimensions. (22%)
SODA: Protecting Proprietary Information in On-Device Machine Learning Models. (4%)
Energy-based learning algorithms for analog computing: a comparative study. (2%)
Adaptive Domain Inference Attack. (1%)
2023-12-21
AutoAugment Input Transformation for Highly Transferable Targeted Attacks. (99%)
Where and How to Attack? A Causality-Inspired Recipe for Generating Counterfactual Adversarial Examples. (98%)
Elevating Defenses: Bridging Adversarial Training and Watermarking for Model Resilience. (86%)
Adversarial Infrared Curves: An Attack on Infrared Pedestrian Detectors in the Physical World. (74%)
Exploiting Novel GPT-4 APIs. (8%)
2023-12-20
Mutual-modality Adversarial Attack with Semantic Perturbation. (99%)
LRS: Enhancing Adversarial Transferability through Lipschitz Regularized Surrogate. (99%)
Adversarial Markov Games: On Adaptive Decision-Based Attacks and Defenses. (98%)
Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models. (98%)
PGN: A perturbation generation network against deep reinforcement learning. (96%)
ARBiBench: Benchmarking Adversarial Robustness of Binarized Neural Networks. (96%)
Scaling Compute Is Not All You Need for Adversarial Robustness. (93%)
Doubly Perturbed Task Free Continual Learning. (9%)
Interactive Visualization of Time-Varying Flow Fields Using Particle Tracing Neural Networks. (1%)
2023-12-19
Tensor Train Decomposition for Adversarial Attacks on Computer Vision Models. (96%)
Rethinking Randomized Smoothing from the Perspective of Scalability. (86%)
SkyMask: Attack-agnostic Robust Federated Learning with Fine-grained Learnable Masks. (74%)
Progressive Poisoned Data Isolation for Training-time Backdoor Defense. (61%)
Adversarial AutoMixup. (11%)
Shaping Up SHAP: Enhancing Stability through Layer-Wise Neighbor Selection. (1%)
I-CEE: Tailoring Explanations of Image Classifications Models to User Expertise. (1%)
2023-12-18
Gemini: A Family of Highly Capable Multimodal Models. (99%)
Adv-Diffusion: Imperceptible Adversarial Face Identity Attack via Latent Diffusion Model. (99%)
The Ultimate Combo: Boosting Adversarial Example Transferability by Composing Data Augmentations. (99%)
DataElixir: Purifying Poisoned Dataset to Mitigate Backdoor Attacks via Diffusion Models. (16%)
A Comprehensive Survey of Attack Techniques, Implementation, and Mitigation Strategies in Large Language Models. (10%)
Model Stealing Attack against Recommender System. (10%)
Model Stealing Attack against Graph Classification with Authenticity, Uncertainty and Diversity. (4%)
MISA: Unveiling the Vulnerabilities in Split Federated Learning. (1%)
A Survey of Side-Channel Attacks in Context of Cache -- Taxonomies, Analysis and Mitigation. (1%)
2023-12-17
UltraClean: A Simple Framework to Train Robust Neural Networks against Backdoor Attacks. (98%)
The Pros and Cons of Adversarial Robustness. (92%)
A Mutation-Based Method for Multi-Modal Jailbreaking Attack Detection. (80%)
Robust Node Representation Learning via Graph Variational Diffusion Networks. (11%)
A Study on Transferability of Deep Learning Models for Network Intrusion Detection. (4%)
2023-12-16
Perturbation-Invariant Adversarial Training for Neural Ranking Models: Improving the Effectiveness-Robustness Trade-Off. (99%)
Rethinking Robustness of Model Attributions. (80%)
SAME: Sample Reconstruction Against Model Extraction Attacks. (13%)
TrojFair: Trojan Fairness Attacks. (8%)
Transformers in Unsupervised Structure-from-Motion. (3%)
TrojFSP: Trojan Insertion in Few-shot Prompt Tuning. (2%)
2023-12-15
LogoStyleFool: Vitiating Video Recognition Systems via Logo Style Transfer. (99%)
Embodied Adversarial Attack: A Dynamic Robust Physical Attack in Autonomous Driving. (99%)
Towards Transferable Targeted 3D Adversarial Attack in the Physical World. (99%)
A Malware Classification Survey on Adversarial Attacks and Defences. (98%)
FlowMur: A Stealthy and Practical Audio Backdoor Attack with Limited Knowledge. (76%)
Closing the Gap: Achieving Better Accuracy-Robustness Tradeoffs Against Query-Based Attacks. (74%)
Fragility, Robustness and Antifragility in Deep Learning. (67%)
VNN: Verification-Friendly Neural Networks with Hard Robustness Guarantees. (67%)
Silent Guardian: Protecting Text from Malicious Exploitation by Large Language Models. (10%)
2023-12-14
AVA: Inconspicuous Attribute Variation-based Adversarial Attack bypassing DeepFake Detection. (99%)
Continual Adversarial Defense. (95%)
SlowTrack: Increasing the Latency of Camera-based Perception in Autonomous Driving Using Adversarial Examples. (92%)
On the Difficulty of Defending Contrastive Learning against Backdoor Attacks. (84%)
Detection and Defense of Unlearnable Examples. (81%)
Improve Robustness of Reinforcement Learning against Observation Perturbations via $l_\infty$ Lipschitz Policy Networks. (81%)
Adversarial Robustness on Image Classification with $k$-means. (81%)
Data and Model Poisoning Backdoor Attacks on Wireless Federated Learning, and the Defense Mechanisms: A Comprehensive Survey. (76%)
DRAM-Locker: A General-Purpose DRAM Protection Mechanism against Adversarial DNN Weight Attacks. (45%)
No-Skim: Towards Efficiency Robustness Evaluation on Skimming-based Language Models. (45%)
Forbidden Facts: An Investigation of Competing Objectives in Llama-2. (45%)
Coevolutionary Algorithm for Building Robust Decision Trees under Minimax Regret. (13%)
Exploring Transferability for Randomized Smoothing. (5%)
Split-Ensemble: Efficient OOD-aware Ensemble via Task and Model Splitting. (1%)
2023-12-13
Defenses in Adversarial Machine Learning: A Survey. (99%)
Robust Few-Shot Named Entity Recognition with Boundary Discrimination and Correlation Purification. (99%)
Universal Adversarial Framework to Improve Adversarial Robustness for Diabetic Retinopathy Detection. (98%)
Towards Inductive Robustness: Distilling and Fostering Wave-induced Resonance in Transductive GCNs Against Graph Adversarial Attacks. (83%)
Scalable Ensemble-based Detection Method against Adversarial Attacks for speaker verification. (64%)
Accelerating the Global Aggregation of Local Explanations. (47%)
Erasing Self-Supervised Learning Backdoor by Cluster Activation Masking. (22%)
Efficient Representation of the Activation Space in Deep Neural Networks. (11%)
Efficient Toxic Content Detection by Bootstrapping and Distilling Large Language Models. (1%)
2023-12-12
Radio Signal Classification by Adversarially Robust Quantum Machine Learning. (99%)
SSTA: Salient Spatially Transformed Attack. (99%)
DTA: Distribution Transform-based Attack for Query-Limited Scenario. (99%)
May the Noise be with you: Adversarial Training without Adversarial Examples. (98%)
Collapse-Oriented Adversarial Training with Triplet Decoupling for Robust Image Retrieval. (98%)
Focus on Hiders: Exploring Hidden Threats for Enhancing Adversarial Training. (98%)
QuadAttack: A Quadratic Programming Approach to Ordered Top-K Attacks. (97%)
Attacking the Loop: Adversarial Attacks on Graph-based Loop Closure Detection. (92%)
ReRoGCRL: Representation-based Robustness in Goal-Conditioned Reinforcement Learning. (86%)
Robust MRI Reconstruction by Smoothed Unrolling (SMUG). (82%)
Cost Aware Untargeted Poisoning Attack against Graph Neural Networks,. (70%)
EdgePruner: Poisoned Edge Pruning in Graph Contrastive Learning. (47%)
Causality Analysis for Evaluating the Security of Large Language Models. (22%)
SimAC: A Simple Anti-Customization Method for Protecting Face Privacy against Text-to-Image Synthesis of Diffusion Models. (13%)
Divide-and-Conquer Attack: Harnessing the Power of LLM to Bypass Safety Filters of Text-to-Image Models. (8%)
Eroding Trust In Aerial Imagery: Comprehensive Analysis and Evaluation Of Adversarial Attacks In Geospatial Systems. (5%)
Securing Graph Neural Networks in MLaaS: A Comprehensive Realization of Query-based Integrity Verification. (2%)
Majority is Not Required: A Rational Analysis of the Private Double-Spend Attack from a Sub-Majority Adversary. (1%)
Rethinking Model Inversion Attacks With Patch-Wise Reconstruction. (1%)
2023-12-11
Towards Transferable Adversarial Attacks with Centralized Perturbation. (99%)
MalPurifier: Enhancing Android Malware Detection with Adversarial Purification against Evasion Attacks. (98%)
Sparse but Strong: Crafting Adversarially Robust Graph Lottery Tickets. (83%)
Reward Certification for Policy Smoothed Reinforcement Learning. (78%)
Activation Gradient based Poisoned Sample Detection Against Backdoor Attacks. (31%)
Poisoned ChatGPT Finds Work for Idle Hands: Exploring Developers' Coding Practices with Insecure Suggestions from Poisoned AI Models. (22%)
Promoting Counterfactual Robustness through Diversity. (13%)
Resilient Path Planning for UAVs in Data Collection under Adversarial Attacks. (10%)
Adversarial Camera Patch: An Effective and Robust Physical-World Attack on Object Detectors. (1%)
Robust Graph Neural Network based on Graph Denoising. (1%)
2023-12-10
Data-Free Hard-Label Robustness Stealing Attack. (86%)
A Practical Survey on Emerging Threats from AI-driven Voice Attacks: How Vulnerable are Commercial Voice Control Systems? (76%)
An Ambiguity Measure for Recognizing the Unknowns in Deep Learning. (12%)
METAL: Metamorphic Testing Framework for Analyzing Large-Language Model Qualities. (2%)
2023-12-09
Poisoning $\times$ Evasion: Symbiotic Adversarial Robustness for Graph Neural Networks. (99%)
Improving Adversarial Robust Fairness via Anti-Bias Soft Label Distillation. (98%)
Dynamic Adversarial Attacks on Autonomous Driving Systems. (98%)
Initialization Matters for Adversarial Transfer Learning. (76%)
2023-12-08
HC-Ref: Hierarchical Constrained Refinement for Robust Adversarial Training of GNNs. (99%)
SA-Attack: Improving Adversarial Transferability of Vision-Language Pre-training Models via Self-Augmentation. (99%)
MIMIR: Masked Image Modeling for Mutual Information-based Adversarial Robustness. (99%)
BELT: Old-School Backdoor Attacks can Evade the State-of-the-Art Defense with Backdoor Exclusivity Lifting. (96%)
An adversarial attack approach for eXplainable AI evaluation on deepfake detection models. (38%)
A Red Teaming Framework for Securing AI in Maritime Autonomous Systems. (3%)
Annotation-Free Group Robustness via Loss-Based Resampling. (2%)
HuRef: HUman-REadable Fingerprint for Large Language Models. (2%)
Topology-Based Reconstruction Prevention for Decentralised Learning. (1%)
2023-12-07
MimicDiffusion: Purifying Adversarial Perturbation via Mimicking Clean Diffusion Model. (99%)
OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization. (99%)
Diffence: Fencing Membership Privacy With Diffusion Models. (97%)
FreqFed: A Frequency Analysis-Based Approach for Mitigating Poisoning Attacks in Federated Learning. (70%)
Forcing Generative Models to Degenerate Ones: The Power of Data Poisoning Attacks. (64%)
DeceptPrompt: Exploiting LLM-driven Code Generation via Adversarial Natural Language Instructions. (15%)
2023-12-06
Defense against ML-based Power Side-channel Attacks on DNN Accelerators with Adversarial Attacks. (98%)
Defense Against Adversarial Attacks using Convolutional Auto-Encoders. (97%)
Node-aware Bi-smoothing: Certified Robustness against Graph Injection Attacks. (88%)
RoAST: Robustifying Language Models via Adversarial Perturbation with Selective Training. (54%)
Detecting Voice Cloning Attacks via Timbre Watermarking. (13%)
Synthesizing Physical Backdoor Datasets: An Automated Framework Leveraging Deep Generative Models. (11%)
Dr. Jekyll and Mr. Hyde: Two Faces of LLMs. (4%)
MICRO: Model-Based Offline Reinforcement Learning with a Conservative Bellman Operator. (2%)
2023-12-05
Generating Visually Realistic Adversarial Patch. (99%)
A Simple Framework to Enhance the Adversarial Robustness of Deep Learning-based Intrusion Detection System. (99%)
Realistic Scatterer Based Adversarial Attacks on SAR Image Classifiers. (99%)
ScAR: Scaling Adversarial Robustness for LiDAR Object Detection. (99%)
Class Incremental Learning for Adversarial Robustness. (98%)
Provable Adversarial Robustness for Group Equivariant Tasks: Graphs, Point Clouds, Molecules, and More. (89%)
On the Robustness of Large Multimodal Models Against Image Adversarial Attacks. (69%)
Scaling Laws for Adversarial Attacks on Language Model Activations. (50%)
Indirect Gradient Matching for Adversarial Robust Distillation. (13%)
Robust Backdoor Detection for Deep Learning via Topological Evolution Dynamics. (3%)
Prompt Optimization via Adversarial In-Context Learning. (3%)
Privacy-Preserving Task-Oriented Semantic Communications Against Model Inversion Attacks. (2%)
Machine Vision Therapy: Multimodal Large Language Models Can Enhance Visual Robustness via Denoising In-Context Learning. (2%)
2023-12-04
Adversarial Medical Image with Hierarchical Feature Hiding. (99%)
InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models. (99%)
Singular Regularization with Information Bottleneck Improves Model's Adversarial Robustness. (98%)
Two-stage optimized unified adversarial patch for attacking visible-infrared cross-modal detectors in the physical world. (12%)
Auto DP-SGD: Dual Improvements of Privacy and Accuracy via Automatic Clipping Threshold and Noise Multiplier Estimation. (1%)
Rejuvenating image-GPT as Strong Visual Representation Learners. (1%)
2023-12-03
QuantAttack: Exploiting Dynamic Quantization to Attack Vision Transformers. (99%)
OCGEC: One-class Graph Embedding Classification for DNN Backdoor Detection. (61%)
Evaluating the Security of Satellite Systems. (16%)
Exploring Adversarial Robustness of LiDAR-Camera Fusion Model in Autonomous Driving. (13%)
Towards Sample-specific Backdoor Attack with Clean Labels via Attribute Trigger. (2%)
2023-12-02
TranSegPGD: Improving Transferability of Adversarial Examples on Semantic Segmentation. (99%)
Rethinking PGD Attack: Is Sign Function Necessary? (98%)
PROFL: A Privacy-Preserving Federated Learning Method with Stringent Defense Against Poisoning Attacks. (61%)
Mendata: A Framework to Purify Manipulated Training Data. (2%)
2023-12-01
PyraTrans: Learning Attention-Enriched Multi-Scale Pyramid Network from Pre-Trained Transformers for Effective Malicious URL Detection. (69%)
Survey of Security Issues in Memristor-based Machine Learning Accelerators for RF Analysis. (22%)
Deep Generative Attacks and Countermeasures for Data-Driven Offline Signature Verification. (10%)
The Philosopher's Stone: Trojaning Plugins of Large Language Models. (4%)
Temperature Balancing, Layer-wise Weight Analysis, and Neural Network Training. (1%)
Crystal: Enhancing Blockchain Mining Transparency with Quorum Certificate. (1%)
2023-11-30
Improving the Robustness of Quantized Deep Neural Networks to White-Box Attacks using Stochastic Quantization and Information-Theoretic Ensemble Training. (98%)
Adversarial Attacks and Defenses for Wireless Signal Classifiers using CDI-aware GANs. (98%)
Universal Backdoor Attacks. (97%)
Fool the Hydra: Adversarial Attacks against Multi-view Object Detection Systems. (97%)
Corrupting Convolution-based Unlearnable Datasets with Pixel-based Image Transformations. (88%)
Optimal Attack and Defense for Reinforcement Learning. (76%)
Can Protective Perturbation Safeguard Personal Data from Being Exploited by Stable Diffusion? (74%)
Improving Adversarial Transferability via Model Alignment. (68%)
Data-Agnostic Model Poisoning against Federated Learning: A Graph Autoencoder Approach. (62%)
Mark My Words: Analyzing and Evaluating Language Model Watermarks. (9%)
2023-11-29
Improving the Robustness of Transformer-based Large Language Models with Dynamic Attention. (98%)
Group-wise Sparse and Explainable Adversarial Attacks. (96%)
Quantum Neural Networks under Depolarization Noise: Exploring White-Box Attacks and Defenses. (88%)
On the Adversarial Robustness of Graph Contrastive Learning Methods. (83%)
Adversarial Robust Memory-Based Continual Learner. (81%)
Improving Faithfulness for Vision Transformers. (80%)
TARGET: Template-Transferable Backdoor Attack Against Prompt-based NLP Models via GPT4. (68%)
Topology-Preserving Adversarial Training. (10%)
Query-Relevant Images Jailbreak Large Multi-Modal Models. (9%)
Analyzing and Explaining Image Classifiers via Diffusion Guidance. (8%)
Poisoning Attacks Against Contrastive Recommender Systems. (2%)
SenTest: Evaluating Robustness of Sentence Encoders. (2%)
Critical Influence of Overparameterization on Sharpness-aware Minimization. (1%)
CLIPC8: Face liveness detection algorithm based on image-text pairs and contrastive learning. (1%)
Unveiling the Implicit Toxicity in Large Language Models. (1%)
2023-11-28
Vulnerability Analysis of Transformer-based Optical Character Recognition to Adversarial Attacks. (99%)
NeRFTAP: Enhancing Transferability of Adversarial Patches on Face Recognition using Neural Radiance Fields. (99%)
Efficient Key-Based Adversarial Defense for ImageNet by Using Pre-trained Model. (98%)
RADAP: A Robust and Adaptive Defense Against Diverse Adversarial Patches on Face Recognition. (92%)
STR-Cert: Robustness Certification for Deep Text Recognition on Deep Learning Pipelines and Vision Transformers. (26%)
1-Lipschitz Layers Compared: Memory, Speed, and Certifiable Robustness. (13%)
Scalable Extraction of Training Data from (Production) Language Models. (10%)
Cooperative Abnormal Node Detection with Adversary Resistance. (10%)
On robust overfitting: adversarial training induced distribution matters. (1%)
Understanding the (Extra-)Ordinary: Validating Deep Model Decisions with Prototypical Concept-based Explanations. (1%)
Shadows Don't Lie and Lines Can't Bend! Generative Models don't know Projective Geometry...for now. (1%)
Enhancing Cyber-Resilience in Integrated Energy System Scheduling with Demand Response Using Deep Reinforcement Learning. (1%)
2023-11-27
RetouchUAA: Unconstrained Adversarial Attack via Image Retouching. (99%)
Adversaral Doodles: Interpretable and Human-drawable Attacks Provide Describable Insights. (99%)
Rethinking Mixup for Improving the Adversarial Transferability. (98%)
Instruct2Attack: Language-Guided Semantic Adversarial Attacks. (98%)
CLAP: Contrastive Learning with Augmented Prompts for Robustness on Pretrained Vision-Language Models. (95%)
A Survey on Vulnerability of Federated Learning: A Learning Algorithm Perspective. (50%)
Threshold Breaker: Can Counter-Based RowHammer Prevention Mechanisms Truly Safeguard DRAM? (31%)
Elijah: Eliminating Backdoors Injected in Diffusion Models via Distribution Shift. (31%)
Distributed Attacks over Federated Reinforcement Learning-enabled Cell Sleep Control. (22%)
"Do Users fall for Real Adversarial Phishing?" Investigating the Human response to Evasive Webpages. (15%)
How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs. (12%)
Microarchitectural Security of AWS Firecracker VMM for Serverless Cloud Platforms. (1%)
2023-11-26
Adversarial Purification of Information Masking. (99%)
Having Second Thoughts? Let's hear it. (56%)
BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIP. (13%)
Confidence Is All You Need for MI Attacks. (2%)
2023-11-25
Mixing Classifiers to Alleviate the Accuracy-Robustness Trade-Off. (68%)
Robust Graph Neural Networks via Unbiased Aggregation. (12%)
Effective Backdoor Mitigation Depends on the Pre-training Objective. (10%)
2023-11-24
Trainwreck: A damaging adversarial attack on image classifiers. (99%)
Segment (Almost) Nothing: Prompt-Agnostic Adversarial Attacks on Segmentation Models. (96%)
Universal Jailbreak Backdoors from Poisoned Human Feedback. (1%)
2023-11-23
When Side-Channel Attacks Break the Black-Box Property of Embedded Artificial Intelligence. (99%)
Adversarial defense based on distribution transfer. (99%)
Robust and Interpretable COVID-19 Diagnosis on Chest X-ray Images using Adversarial Training. (68%)
Presentation Attack Detection using Convolutional Neural Networks and Local Binary Patterns. (1%)
2023-11-22
A Survey of Adversarial CAPTCHAs on its History, Classification and Generation. (99%)
Transfer Attacks and Defenses for Large Language Models on Coding Tasks. (99%)
Panda or not Panda? Understanding Adversarial Attacks with Interactive Visualization. (98%)
Hard Label Black Box Node Injection Attack on Graph Neural Networks. (93%)
Security and Privacy Challenges in Deep Learning Models. (74%)
A Somewhat Robust Image Watermark against Diffusion-based Editing Models. (50%)
OASIS: Offsetting Active Reconstruction Attacks in Federated Learning. (15%)
Unified Classification and Rejection: A One-versus-All Framework. (1%)
2023-11-21
SD-NAE: Generating Natural Adversarial Examples with Stable Diffusion. (96%)
Stable Unlearnable Example: Enhancing the Robustness of Unlearnable Examples via Stable Error-Minimizing Noise. (96%)
Attention Deficit is Ordered! Fooling Deformable Vision Transformers with Collaborative Adversarial Patches. (75%)
Attacking Motion Planners Using Adversarial Perception Errors. (69%)
Toward Robust Imperceptible Perturbation against Unauthorized Text-to-image Diffusion-based Synthesis. (62%)
Iris Presentation Attack: Assessing the Impact of Combining Vanadium Dioxide Films with Artificial Eyes. (1%)
2023-11-20
ODDR: Outlier Detection & Dimension Reduction Based Defense Against Adversarial Patches. (99%)
DefensiveDR: Defending against Adversarial Patches using Dimensionality Reduction. (99%)
Generating Valid and Natural Adversarial Examples with Large Language Models. (99%)
AdvGen: Physical Adversarial Attack on Face Presentation Attack Detection Systems. (99%)
Beyond Boundaries: A Comprehensive Survey of Transferable Attacks on AI Systems. (50%)
Understanding Variation in Subpopulation Susceptibility to Poisoning Attacks. (15%)
Training robust and generalizable quantum models. (10%)
BrainWash: A Poisoning Attack to Forget in Continual Learning. (4%)
2023-11-19
Adversarial Prompt Tuning for Vision-Language Models. (98%)
Token-Level Adversarial Prompt Detection Based on Perplexity Measures and Contextual Information. (78%)
BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning. (69%)
EditShield: Protecting Unauthorized Image Editing by Instruction-guided Diffusion Models. (10%)
2023-11-18
Boost Adversarial Transferability by Uniform Scale and Mix Mask Method. (99%)
Improving Adversarial Transferability by Stable Diffusion. (99%)
Attention-Based Real-Time Defenses for Physical Adversarial Attacks in Vision Applications. (92%)
TextGuard: Provable Defense against Backdoor Attacks on Text Classification. (82%)
Robust Network Slicing: Multi-Agent Policies, Adversarial Attacks, and Defensive Strategies. (1%)
2023-11-17
Breaking Temporal Consistency: Generating Video Universal Adversarial Perturbations Using Image Models. (97%)
PACOL: Poisoning Attacks Against Continual Learners. (93%)
Two-Factor Authentication Approach Based on Behavior Patterns for Defeating Puppet Attacks. (1%)
2023-11-16
Breaking Boundaries: Balancing Performance and Robustness in Deep Wireless Traffic Forecasting. (99%)
Hijacking Large Language Models via Adversarial In-Context Learning. (92%)
Cognitive Overload: Jailbreaking Large Language Models with Overloaded Logical Thinking. (54%)
Test-time Backdoor Mitigation for Black-Box Large Language Models with Defensive Demonstrations. (38%)
RLHFPoison: Reward Poisoning Attack for Reinforcement Learning with Human Feedback in Large Language Models. (16%)
Towards Improving Robustness Against Common Corruptions using Mixture of Class Specific Experts. (2%)
Understanding the Effectiveness of Large Language Models in Detecting Security Vulnerabilities. (2%)
Bergeron: Combating Adversarial Attacks through a Conscience-Based Alignment Framework. (2%)
Towards more Practical Threat Models in Artificial Intelligence Security. (2%)
You Cannot Escape Me: Detecting Evasions of SIEM Rules in Enterprise Networks. (1%)
2023-11-15
Jailbreaking GPT-4V via Self-Adversarial Attacks with System Prompts. (99%)
Backdoor Activation Attack: Attack Large Language Models using Activation Steering for Safety-Alignment. (74%)
Fast Certification of Vision-Language Models Using Incremental Randomized Smoothing. (64%)
Adversarially Robust Spiking Neural Networks Through Conversion. (61%)
How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilities. (16%)
Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization. (16%)
Privacy Threats in Stable Diffusion Models. (13%)
MirrorNet: A TEE-Friendly Framework for Secure On-device DNN Inference. (2%)
JAB: Joint Adversarial Prompting and Belief Augmentation. (1%)
Beyond Detection: Unveiling Fairness Vulnerabilities in Abusive Language Models. (1%)
2023-11-14
Towards Improving Robustness Against Common Corruptions in Object Detectors Using Adversarial Contrastive Learning. (99%)
Physical Adversarial Examples for Multi-Camera Systems. (99%)
DALA: A Distribution-Aware LoRA-Based Adversarial Attack against Language Models. (99%)
On The Relationship Between Universal Adversarial Attacks And Sparse Representations. (98%)
A Wolf in Sheep's Clothing: Generalized Nested Jailbreak Prompts can Fool Large Language Models Easily. (62%)
Evaluating Concurrent Robustness of Language Models Across Diverse Challenge Sets. (26%)
The Perception-Robustness Tradeoff in Deterministic Image Restoration. (1%)
2023-11-13
Adversarial Purification for Data-Driven Power System Event Classifiers with Diffusion Models. (99%)
Parrot-Trained Adversarial Examples: Pushing the Practicality of Black-Box Audio Attacks against Speaker Recognition Models. (99%)
An Extensive Study on Adversarial Attack against Pre-trained Models of Code. (99%)
Multi-agent Attacks for Black-box Social Recommendations. (96%)
On the Robustness of Neural Collapse and the Neural Collapse of Robustness. (87%)
Tabdoor: Backdoor Vulnerabilities in Transformer-based Neural Networks for Tabular Data. (70%)
2023-11-12
Learning Globally Optimized Language Structure via Adversarial Training. (83%)
Resilient Graph Neural Networks: A Coupled Dynamical Systems Approach. (70%)
Analytical Verification of Deep Neural Network Performance for Time-Synchronized Distribution System State Estimation. (5%)
DialMAT: Dialogue-Enabled Transformer with Moment-Based Adversarial Training. (1%)
2023-11-11
Robust Text Classification: Analyzing Prototype-Based Networks. (97%)
2023-11-10
Robust Adversarial Attacks Detection for Deep Learning based Relative Pose Estimation for Space Rendezvous. (99%)
Fight Fire with Fire: Combating Adversarial Patch Attacks using Pattern-randomized Defensive Patches. (99%)
Transferability Bound Theory: Exploring Relationship between Adversarial Transferability and Flatness. (99%)
Resilient and constrained consensus against adversarial attacks: A distributed MPC framework. (84%)
CALLOC: Curriculum Adversarial Learning for Secure and Robust Indoor Localization. (1%)
Practical Membership Inference Attacks against Fine-tuned Large Language Models via Self-prompt Calibration. (1%)
2023-11-09
ABIGX: A Unified Framework for eXplainable Fault Detection and Classification. (68%)
Honest Score Client Selection Scheme: Preventing Federated Learning Label Flipping Attacks in Non-IID Scenarios. (50%)
Scale-MIA: A Scalable Model Inversion Attack against Secure Federated Learning via Latent Space Reconstruction. (15%)
FigStep: Jailbreaking Large Vision-language Models via Typographic Visual Prompts. (1%)
FireMatch: A Semi-Supervised Video Fire Detection Network Based on Consistency and Distribution Alignment. (1%)
2023-11-08
Constrained Adaptive Attacks: Realistic Evaluation of Adversarial Examples and Robust Training of Deep Neural Networks for Tabular Data. (99%)
Army of Thieves: Enhancing Black-Box Model Extraction via Ensemble based sample selection. (70%)
Frontier Language Models are not Robust to Adversarial Arithmetic, or "What do I need to say so you agree 2+2=5? (61%)
SCAAT: Improving Neural Network Interpretability via Saliency Constrained Adaptive Adversarial Training. (10%)
Domain Adaptive Object Detection via Balancing Between Self-Training and Adversarial Learning. (1%)
Counter-Empirical Attacking based on Adversarial Reinforcement Learning for Time-Relevant Scoring System. (1%)
2023-11-07
Unveiling Safety Vulnerabilities of Large Language Models. (61%)
When Fairness Meets Privacy: Exploring Privacy Threats in Fair Binary Classifiers through Membership Inference Attacks. (10%)
Identifying and Mitigating Vulnerabilities in LLM-Integrated Applications. (2%)
SoK: Security Below the OS -- A Security Analysis of UEFI. (1%)
Do LLMs exhibit human-like response biases? A case study in survey design. (1%)
2023-11-06
Measuring Adversarial Datasets. (92%)
Can LLMs Follow Simple Rules? (68%)
Preserving Privacy in GANs Against Membership Inference Attack. (33%)
Cal-DETR: Calibrated Detection Transformer. (4%)
2023-11-05
ELEGANT: Certified Defense on the Fairness of Graph Neural Networks. (10%)
2023-11-04
From Trojan Horses to Castle Walls: Unveiling Bilateral Data Poisoning Effects in Diffusion Models. (74%)
2023-11-03
Efficient Black-Box Adversarial Attacks on Neural Text Detectors. (22%)
The Alignment Problem in Context. (2%)
2023-11-02
Adversary ML Resilience in Autonomous Driving Through Human Centered Perception Mechanisms. (99%)
Towards Evaluating Transfer-based Attacks Systematically, Practically, and Fairly. (99%)
Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game. (93%)
On the Lipschitz constant of random neural networks. (92%)
Universal Perturbation-based Secret Key-Controlled Data Hiding. (80%)
Distilling Out-of-Distribution Robustness from Vision-Language Foundation Models. (76%)
Assist Is Just as Important as the Goal: Image Resurfacing to Aid Model's Robust Prediction. (13%)
Robust Adversarial Reinforcement Learning via Bounded Rationality Curricula. (12%)
Sequential Subset Matching for Dataset Distillation. (1%)
E(2) Equivariant Neural Networks for Robust Galaxy Morphology Classification. (1%)
Robust Identity Perceptual Watermark Against Deepfake Face Swapping. (1%)
2023-11-01
NEO-KD: Knowledge-Distillation-Based Adversarial Training for Robust Multi-Exit Neural Networks. (99%)
Adversarial Examples in the Physical World: A Survey. (98%)
Optimal Cost Constrained Adversarial Attacks For Multiple Agent Systems. (80%)
Improving Robustness for Vision Transformer with a Simple Dynamic Scanning Augmentation. (76%)
MIST: Defending Against Membership Inference Attacks Through Membership-Invariant Subspace Training. (75%)
Robustness Tests for Automatic Machine Translation Metrics with Adversarial Attacks. (1%)
Open-Set Face Recognition with Maximal Entropy and Objectosphere Loss. (1%)
2023-10-31
Amoeba: Circumventing ML-supported Network Censorship via Adversarial Reinforcement Learning. (99%)
Robust Safety Classifier for Large Language Models: Adversarial Prompt Shield. (99%)
LFAA: Crafting Transferable Targeted Adversarial Examples with Low-Frequency Perturbations. (99%)
Magmaw: Modality-Agnostic Adversarial Attacks on Machine Learning-Based Wireless Communication Systems. (98%)
Is Robustness Transferable across Languages in Multilingual Neural Machine Translation? (26%)
Dynamic Batch Norm Statistics Update for Natural Robustness. (22%)
In Search of Lost Online Test-time Adaptation: A Survey. (1%)
2023-10-30
Label-Only Model Inversion Attacks via Knowledge Transfer. (83%)
Exploring Geometry of Blind Spots in Vision Models. (83%)
Adversarial Attacks and Defenses in Large Language Models: Old and New Threats. (74%)
Generated Distributions Are All You Need for Membership Inference Attacks Against Generative Models. (61%)
Causal Fair Metric: Bridging Causality, Individual Fairness, and Adversarial Robustness. (33%)
Differentially Private Reward Estimation with Preference Feedback. (16%)
Asymmetric Diffusion Based Channel-Adaptive Secure Wireless Semantic Communications. (10%)
Privacy-Preserving Federated Learning over Vertically and Horizontally Partitioned Data for Financial Anomaly Detection. (1%)
2023-10-29
Blacksmith: Fast Adversarial Training of Vision Transformers via a Mixture of Single-step and Multi-step Methods. (99%)
Boosting Decision-Based Black-Box Adversarial Attack with Gradient Priors. (98%)
BERT Lost Patience Won't Be Robust to Adversarial Slowdown. (98%)
Adversarial Examples Are Not Real Features. (98%)
IMPRESS: Evaluating the Resilience of Imperceptible Perturbations Against Unauthorized Data Usage in Diffusion-Based Generative AI. (82%)
Poisoning Retrieval Corpora by Injecting Adversarial Passages. (68%)
Label Poisoning is All You Need. (54%)
Robustifying Language Models with Test-Time Adaptation. (47%)
Path Analysis for Effective Fault Localization in Deep Neural Networks. (1%)
From Chatbots to PhishBots? -- Preventing Phishing scams created using ChatGPT, Google Bard and Claude. (1%)
2023-10-28
Assessing and Improving Syntactic Adversarial Robustness of Pre-trained Models for Code Translation. (92%)
Benchmark Generation Framework with Customizable Distortions for Image Classifier Robustness. (86%)
Purify++: Improving Diffusion-Purification with Advanced Diffusion Models and Control of Randomness. (61%)
Large Language Models Are Better Adversaries: Exploring Generative Clean-Label Backdoor Attacks Against Text Classifiers. (47%)
Where have you been? A Study of Privacy Risk for Point-of-Interest Recommendation. (10%)
2023-10-27
DiffAttack: Evasion Attacks Against Diffusion-Based Adversarial Purification. (99%)
Understanding and Improving Ensemble Adversarial Defense. (99%)
LipSim: A Provably Robust Perceptual Similarity Metric. (45%)
Elevating Code-mixed Text Handling through Auditory Information of Words. (5%)
Understanding Parameter Saliency via Extreme Value Theory. (1%)
2023-10-26
Unscrambling the Rectification of Adversarial Attacks Transferability across Computer Networks. (99%)
A Survey on Transferability of Adversarial Examples across Deep Neural Networks. (99%)
Defending Against Transfer Attacks From Public Models. (99%)
Uncertainty-weighted Loss Functions for Improved Adversarial Attacks on Semantic Segmentation. (93%)
Detection Defenses: An Empty Promise against Adversarial Patch Attacks on Optical Flow. (93%)
CBD: A Certified Backdoor Detector Based on Local Dominant Probability. (76%)
SoK: Pitfalls in Evaluating Black-Box Attacks. (76%)
Instability of computer vision models is a necessary result of the task itself. (26%)
PAC-tuning:Fine-tuning Pretrained Language Models with PAC-driven Perturbed Gradient Descent. (1%)
A minimax optimal control approach for robust neural ODEs. (1%)
2023-10-25
Break it, Imitate it, Fix it: Robustness by Generating Human-Like Attacks. (93%)
Trust, but Verify: Robust Image Segmentation using Deep Learning. (54%)
Dual Defense: Adversarial, Traceable, and Invisible Robust Watermarking against Face Swapping. (26%)
On the Proactive Generation of Unsafe Images From Text-To-Image Models Using Benign Prompts. (22%)
Wide Flat Minimum Watermarking for Robust Ownership Verification of GANs. (12%)
Multi-scale Diffusion Denoised Smoothing. (1%)
SparseDFF: Sparse-View Feature Distillation for One-Shot Dexterous Manipulation. (1%)
2023-10-24
Adversarial sample generation and training using geometric masks for accurate and resilient license plate character recognition. (99%)
RAEDiff: Denoising Diffusion Probabilistic Models Based Reversible Adversarial Examples Self-Generation and Self-Recovery. (92%)
Defense Against Model Extraction Attacks on Recommender Systems. (92%)
Segue: Side-information Guided Generative Unlearnable Examples for Facial Privacy Protection in Real World. (89%)
Hierarchical Randomized Smoothing. (75%)
Momentum Gradient-based Untargeted Attack on Hypergraph Neural Networks. (73%)
Corrupting Neuron Explanations of Deep Visual Features. (41%)
Improving Robustness and Reliability in Medical Image Classification with Latent-Guided Diffusion and Nested-Ensembles. (13%)
Guiding LLM to Fool Itself: Automatically Manipulating Machine Reading Comprehension Shortcut Triggers. (10%)
A Survey on Detection of LLMs-Generated Content. (1%)
White-box Compiler Fuzzing Empowered by Large Language Models. (1%)
Enhancing Large Language Models for Secure Code Generation: A Dataset-driven Study on Vulnerability Mitigation. (1%)
2023-10-23
Semantic-Aware Adversarial Training for Reliable Deep Hashing Retrieval. (99%)
F$^2$AT: Feature-Focusing Adversarial Training via Disentanglement of Natural and Perturbed Patterns. (99%)
AutoDAN: Automatic and Interpretable Adversarial Attacks on Large Language Models. (98%)
Fast Propagation is Better: Accelerating Single-Step Adversarial Training via Sampling Subnetworks. (98%)
On the Detection of Image-Scaling Attacks in Machine Learning. (15%)
Unleashing the potential of prompt engineering: a comprehensive review. (1%)
RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptions. (1%)
Calibration of Time-Series Forecasting: Detecting and Adapting Context-Driven Distribution Shift. (1%)
The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks. (1%)
2023-10-22
Diffusion-Based Adversarial Purification for Speaker Verification. (99%)
CT-GAT: Cross-Task Generative Adversarial Attack based on Transferability. (99%)
Imperceptible CMOS camera dazzle for adversarial attacks on deep neural networks. (92%)
ADoPT: LiDAR Spoofing Attack Detection Based on Point-Level Temporal Consistency. (26%)
Attention-Enhancing Backdoor Attacks Against BERT-based Models. (13%)
MoPe: Model Perturbation-based Privacy Attacks on Language Models. (9%)
Reputation-Based Federated Learning Defense to Mitigate Threats in EEG Signal Classification. (1%)
2023-10-21
Adversarial Image Generation by Spatial Transformation in Perceptual Colorspaces. (99%)
Training Image Derivatives: Increased Accuracy and Universal Robustness. (5%)
2023-10-20
Beyond Hard Samples: Robust and Effective Grammatical Error Correction with Cycle Self-Augmenting. (99%)
An LLM can Fool Itself: A Prompt-Based Adversarial Attack. (99%)
Prompt-Specific Poisoning Attacks on Text-to-Image Generative Models. (61%)
The Hidden Adversarial Vulnerabilities of Medical Federated Learning. (45%)
Adversarial Attacks on Fairness of Graph Neural Networks. (26%)
FLTracer: Accurate Poisoning Attack Provenance in Federated Learning. (26%)
Can We Trust the Similarity Measurement in Federated Learning? (15%)
Data-Free Knowledge Distillation Using Adversarially Perturbed OpenGL Shader Images. (4%)
VOICE-ZEUS: Impersonating Zoom's E2EE-Protected Static Media and Textual Communications via Simple Voice Manipulations. (4%)
2023-10-19
Automatic Hallucination Assessment for Aligned Large Language Models via Transferable Adversarial Attacks. (98%)
Generating Robust Adversarial Examples against Online Social Networks (OSNs). (98%)
Recoverable Privacy-Preserving Image Classification through Noise-like Adversarial Examples. (98%)
Learn from the Past: A Proxy based Adversarial Defense Framework to Boost Robustness. (98%)
OODRobustBench: benchmarking and analyzing adversarial robustness under distribution shift. (97%)
PatchCURE: Improving Certifiable Robustness, Model Utility, and Computation Efficiency of Adversarial Patch Defenses. (97%)
Prompt Injection Attacks and Defenses in LLM-Integrated Applications. (47%)
Attack Prompt Generation for Red Teaming and Defending Large Language Models. (15%)
SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models. (5%)
To grok or not to grok: Disentangling generalization and memorization on corrupted algorithmic datasets. (1%)
Detecting Shared Data Manipulation in Distributed Optimization Algorithms. (1%)
Towards Robust Pruning: An Adaptive Knowledge-Retention Pruning Strategy for Language Models. (1%)
2023-10-18
Exploring Decision-based Black-box Attacks on Face Forgery Detection. (99%)
Segment Anything Meets Universal Adversarial Perturbation. (99%)
IRAD: Implicit Representation-driven Image Resampling against Adversarial Attacks. (99%)
Revisiting Transferable Adversarial Image Examples: Attack Categorization, Evaluation Guidelines, and New Insights. (99%)
Tailoring Adversarial Attacks on Deep Neural Networks for Targeted Class Manipulation Using DeepFool Algorithm. (99%)
Malicious Agent Detection for Robust Multi-Agent Collaborative Perception. (87%)
Black-Box Training Data Identification in GANs via Detector Networks. (82%)
Adversarial Training for Physics-Informed Neural Networks. (81%)
REVAMP: Automated Simulations of Adversarial Attacks on Arbitrary Objects in Realistic Scenes. (80%)
Quantifying Privacy Risks of Prompts in Visual Prompt Learning. (76%)
To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now. (47%)
CAT: Closed-loop Adversarial Training for Safe End-to-End Driving. (2%)
PrivInfer: Privacy-Preserving Inference for Black-box Large Language Model. (1%)
2023-10-17
The Efficacy of Transformer-based Adversarial Attacks in Security Domains. (99%)
Adversarial Robustness Unhardening via Backdoor Attacks in Federated Learning. (93%)
WaveAttack: Asymmetric Frequency Obfuscation-based Backdoor Attacks Against Deep Neural Networks. (15%)
Generalizability of CNN Architectures for Face Morph Presentation Attack. (1%)
2023-10-16
Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks. (98%)
Regularization properties of adversarially-trained linear regression. (92%)
Fast Adversarial Label-Flipping Attack on Tabular Data. (84%)
A Non-monotonic Smooth Activation Function. (83%)
Quantifying Assistive Robustness Via the Natural-Adversarial Frontier. (68%)
A Comprehensive Study of Privacy Risks in Curriculum Learning. (67%)
DANAA: Towards transferable attacks with double adversarial neuron attribution. (26%)
Demystifying Poisoning Backdoor Attacks from a Statistical Perspective. (9%)
Prompt Packer: Deceiving LLMs through Compositional Instruction with Hidden Attacks. (4%)
Robust Multi-Agent Reinforcement Learning via Adversarial Regularization: Theoretical Foundation and Stable Algorithms. (3%)
Passive Inference Attacks on Split Learning via Adversarial Regularization. (3%)
On the Transferability of Learning Models for Semantic Segmentation for Remote Sensing Data. (2%)
Orthogonal Uncertainty Representation of Data Manifold for Robust Long-Tailed Learning. (1%)
Will the Prince Get True Love's Kiss? On the Model Sensitivity to Gender Perturbation over Fairytale Texts. (1%)
2023-10-15
Towards Deep Learning Models Resistant to Transfer-based Adversarial Attacks via Data-centric Robust Learning. (99%)
SCME: A Self-Contrastive Method for Data-free and Query-Limited Model Extraction Attack. (99%)
AFLOW: Developing Adversarial Examples under Extremely Noise-limited Settings. (99%)
Black-box Targeted Adversarial Attack on Segment Anything (SAM). (99%)
Evading Detection Actively: Toward Anti-Forensics against Forgery Localization. (97%)
Explore the Effect of Data Selection on Poison Efficiency in Backdoor Attacks. (61%)
Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models? (9%)
VFLAIR: A Research Library and Benchmark for Vertical Federated Learning. (3%)
2023-10-14
BufferSearch: Generating Black-Box Adversarial Texts With Lower Queries. (98%)
2023-10-13
Is Certifying $\ell_p$ Robustness Still Worthwhile? (99%)
User Inference Attacks on Large Language Models. (41%)
On the Over-Memorization During Natural, Robust and Catastrophic Overfitting. (1%)
2023-10-12
Samples on Thin Ice: Re-Evaluating Adversarial Pruning of Neural Networks. (99%)
Concealed Electronic Countermeasures of Radar Signal with Adversarial Examples. (93%)
Attacks Meet Interpretability (AmI) Evaluation and Findings. (92%)
Provably Robust Cost-Sensitive Learning via Randomized Smoothing. (73%)
Improving Fast Minimum-Norm Attacks with Hyperparameter Optimization. (68%)
Fed-Safe: Securing Federated Learning in Healthcare Against Adversarial Attacks. (64%)
Bucks for Buckets (B4B): Active Defenses Against Stealing Encoders. (31%)
Sentinel: An Aggregation Function to Secure Decentralized Federated Learning. (13%)
Defending Our Privacy With Backdoors. (10%)
Investigating the Robustness and Properties of Detection Transformers (DETR) Toward Difficult Images. (9%)
Polynomial Time Cryptanalytic Extraction of Neural Network Models. (3%)
SEE-OoD: Supervised Exploration For Enhanced Out-of-Distribution Detection. (1%)
XAI Benchmark for Visual Explanation. (1%)
Jailbreaking Black Box Large Language Models in Twenty Queries. (1%)
Voyager: MTD-Based Aggregation Protocol for Mitigating Poisoning Attacks on DFL. (1%)
2023-10-11
Boosting Black-box Attack to Deep Neural Networks with Conditional Diffusion Models. (99%)
Promoting Robustness of Randomized Smoothing: Two Cost-Effective Approaches. (89%)
An Adversarial Example for Direct Logit Attribution: Memory Management in gelu-4l. (13%)
Prompt Backdoors in Visual Prompt Learning. (11%)
Why Train More? Effective and Efficient Membership Inference via Memorization. (10%)
Towards Causal Deep Learning for Vulnerability Detection. (4%)
Deep Reinforcement Learning for Autonomous Cyber Defence: A Survey. (4%)
2023-10-10
A Geometrical Approach to Evaluate the Adversarial Robustness of Deep Neural Networks. (99%)
My Brother Helps Me: Node Injection Based Adversarial Attack on Social Bot Detection. (98%)
Adversarial Robustness in Graph Neural Networks: A Hamiltonian Approach. (83%)
Adversarial optimization leads to over-optimistic security-constrained dispatch, but sampling can help. (76%)
No Privacy Left Outside: On the (In-)Security of TEE-Shielded DNN Partition for On-Device ML. (62%)
Comparing the Robustness of Modern No-Reference Image- and Video-Quality Metrics to Adversarial Attacks. (47%)
GraphCloak: Safeguarding Task-specific Knowledge within Graph-structured Data from Unauthorized Exploitation. (22%)
Latent Diffusion Counterfactual Explanations. (5%)
FTFT: efficient and robust Fine-Tuning by transFerring Training dynamics. (2%)
Investigating the Adversarial Robustness of Density Estimation Using the Probability Flow ODE. (2%)
Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations. (1%)
2023-10-09
PAC-Bayesian Spectrally-Normalized Bounds for Adversarially Robust Generalization. (92%)
Domain Watermark: Effective and Harmless Dataset Copyright Protection is Closed at Hand. (22%)
Theoretical Analysis of Robust Overfitting for Wide DNNs: An NTK Approach. (5%)
Exploring adversarial attacks in federated learning for medical imaging. (2%)
2023-10-08
GReAT: A Graph Regularized Adversarial Training Method. (99%)
An Initial Investigation of Neural Replay Simulator for Over-the-Air Adversarial Perturbations to Automatic Speaker Verification. (99%)
AdvSV: An Over-the-Air Adversarial Attack Dataset for Speaker Verification. (96%)
Transferable Availability Poisoning Attacks. (83%)
BRAINTEASER: Lateral Thinking Puzzles for Large Language Models. (26%)
Stealthy Backdoor Attack via Confidence-driven Sampling. (10%)
Adversarial Attacks on Combinatorial Multi-Armed Bandits. (5%)
2023-10-07
Improving Adversarial Attacks on Latent Diffusion Model. (99%)
IPMix: Label-Preserving Data Augmentation Method for Training Robust Classifiers. (76%)
Test-Time Adaptation Induces Stronger Accuracy and Agreement-on-the-Line. (1%)
2023-10-06
VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models. (98%)
Generating Less Certain Adversarial Examples Improves Robust Generalization. (98%)
Kick Bad Guys Out! Conditionally Activated Anomaly Detection in Federated Learning with Zero-Knowledge Proof Verification. (84%)
2023-10-05
OMG-ATTACK: Self-Supervised On-Manifold Generation of Transferable Evasion Attacks. (99%)
Untargeted White-box Adversarial Attack with Heuristic Defence Methods in Real-time Deep Learning based Network Intrusion Detection System. (99%)
Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria. (99%)
An Integrated Algorithm for Robust and Imperceptible Audio Adversarial Examples. (98%)
Adversarial Machine Learning for Social Good: Reframing the Adversary as an Ally. (98%)
SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks. (92%)
Better Safe than Sorry: Pre-training CLIP against Targeted Data Poisoning and Backdoor Attacks. (64%)
Targeted Adversarial Attacks on Generalizable Neural Radiance Fields. (56%)
Certification of Deep Learning Models for Medical Image Segmentation. (15%)
Certifiably Robust Graph Contrastive Learning. (5%)
Towards Robust and Generalizable Training: An Empirical Study of Noisy Slot Filling for Input Perturbations. (2%)
2023-10-04
Optimizing Key-Selection for Face-based One-Time Biometrics via Morphing. (98%)
Misusing Tools in Large Language Models With Visual Adversarial Examples. (97%)
Burning the Adversarial Bridges: Robust Windows Malware Detection Against Binary-level Mutations. (82%)
Raze to the Ground: Query-Efficient Adversarial HTML Attacks on Machine-Learning Phishing Webpage Detectors. (81%)
Shielding the Unseen: Privacy Protection through Poisoning NeRF with Spatial Deformation. (10%)
2023-10-03
Splitting the Difference on Adversarial Training. (99%)
DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training. (97%)
SlowFormer: Universal Adversarial Patch for Attack on Compute and Energy Efficiency of Inference Efficient Vision Transformers. (86%)
Towards Stable Backdoor Purification through Feature Shift Tuning. (83%)
Jailbreaker in Jail: Moving Target Defense for Large Language Models. (73%)
AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models. (56%)
Beyond Labeling Oracles: What does it mean to steal ML models? (47%)
A Recipe for Improved Certifiable Robustness. (22%)
Exploring Model Learning Heterogeneity for Boosting Ensemble Robustness. (13%)
FLEDGE: Ledger-based Federated Learning Resilient to Inference and Backdoor Attacks. (11%)
AutoLoRa: A Parameter-Free Automated Robust Fine-Tuning Framework. (3%)
2023-10-02
Fooling the Textual Fooler via Randomizing Latent Representations. (99%)
LLM Lies: Hallucinations are not Bugs, but Features as Adversarial Examples. (93%)
Adversarial Client Detection via Non-parametric Subspace Monitoring in the Internet of Federated Things. (92%)
LoFT: Local Proxy Fine-tuning For Improving Transferability Of Adversarial Attacks Against Large Language Model. (87%)
Gotcha! This Model Uses My Code! Evaluating Membership Leakage Risks in Code Models. (13%)
Toward effective protection against diffusion based mimicry through score distillation. (3%)
Fool Your (Vision and) Language Model With Embarrassingly Simple Permutations. (1%)
2023-10-01
A Survey of Robustness and Safety of 2D and 3D Deep Learning Models Against Adversarial Attacks. (99%)
Counterfactual Image Generation for adversarially robust and interpretable Classifiers. (96%)
Understanding Adversarial Transferability in Federated Learning. (93%)
On the Onset of Robust Overfitting in Adversarial Training. (87%)
GhostEncoder: Stealthy Backdoor Attacks with Dynamic Triggers to Pre-trained Encoders in Self-supervised Learning. (61%)
Fewer is More: Trojan Attacks on Parameter-Efficient Fine-Tuning. (9%)
Can Pre-trained Networks Detect Familiar Out-of-Distribution Data? (1%)
How well does LLM generate security tests? (1%)
2023-09-30
Understanding the Robustness of Randomized Feature Defense Against Query-Based Adversarial Attacks. (99%)
Human-Producible Adversarial Examples. (98%)
Black-box Attacks on Image Activity Prediction and its Natural Language Explanations. (98%)
Horizontal Class Backdoor to Deep Learning. (84%)
Refutation of Shapley Values for XAI -- Additional Evidence. (8%)
2023-09-29
Robustness of AI-Image Detectors: Fundamental Limits and Practical Attacks. (99%)
Efficient Biologically Plausible Adversarial Training. (98%)
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks. (96%)
On Continuity of Robust and Accurate Classifiers. (93%)
Adversarial Machine Learning in Latent Representations of Neural Networks. (93%)
Certified Robustness via Dynamic Margin Maximization and Improved Lipschitz Regularization. (92%)
Toward Robust Recommendation via Real-time Vicinal Defense. (82%)
Adversarial Explainability: Utilizing Explainable Machine Learning in Bypassing IoT Botnet Detection Systems. (31%)
Practical Membership Inference Attacks Against Large-Scale Multi-Modal Models: A Pilot Study. (13%)
Distributed Resilient Control of DC Microgrids Under Generally Unbounded FDI Attacks. (1%)
Source Inference Attacks: Beyond Membership Inference Attacks in Federated Learning. (1%)
2023-09-28
Investigating Human-Identifiable Features Hidden in Adversarial Perturbations. (98%)
Parameter-Saving Adversarial Training: Reinforcing Multi-Perturbation Robustness via Hypernetworks. (98%)
Towards Poisoning Fair Representations. (70%)
On the Trade-offs between Adversarial Robustness and Actionable Explanations. (68%)
The Lipschitz-Variance-Margin Tradeoff for Enhanced Randomized Smoothing. (56%)
Post-Training Overfitting Mitigation in DNN Classifiers. (41%)
Leveraging Optimization for Adaptive Attacks on Image Watermarks. (13%)
Random and Safe Cache Architecture to Defeat Cache Timing Attacks. (9%)
Robust Offline Reinforcement Learning -- Certify the Confidence Interval. (4%)
A Primer on Bayesian Neural Networks: Review and Debates. (2%)
2023-09-27
On the Computational Entanglement of Distant Features in Adversarial Machine Learning. (99%)
Adversarial Examples Might be Avoidable: The Role of Data Concentration in Adversarial Robustness. (95%)
Defending Against Physical Adversarial Patch Attacks on Infrared Human Detection. (95%)
Automatic Feature Fairness in Recommendation via Adversaries. (33%)
Warfare:Breaking the Watermark Protection of AI-Generated Content. (12%)
Generating Transferable Adversarial Simulation Scenarios for Self-Driving via Neural Rendering. (11%)
Breaking On-Chip Communication Anonymity using Flow Correlation Attacks. (4%)
Genetic Algorithm-Based Dynamic Backdoor Attack on Federated Learning-Based Network Traffic Classification. (1%)
2023-09-26
Structure Invariant Transformation for better Adversarial Transferability. (99%)
Privacy-preserving and Privacy-attacking Approaches for Speech and Audio -- A Survey. (16%)
Neural Stochastic Differential Equations for Robust and Explainable Analysis of Electromagnetic Unintended Radiated Emissions. (2%)
Collaborative Watermarking for Adversarial Speech Synthesis. (1%)
2023-09-25
DifAttack: Query-Efficient Black-Box Attack via Disentangled Feature Space. (99%)
Gray-box Adversarial Attack of Deep Reinforcement Learning-based Trading Agents. (98%)
SurrogatePrompt: Bypassing the Safety Filter of Text-To-Image Models via Substitution. (1%)
2023-09-24
Adversarial Attacks on Video Object Segmentation with Hard Region Discovery. (99%)
Vulnerabilities in Video Quality Assessment Models: The Challenge of Adversarial Attacks. (98%)
On the Effectiveness of Adversarial Samples against Ensemble Learning-based Windows PE Malware Detectors. (86%)
Benchmarking Local Robustness of High-Accuracy Binary Neural Networks for Enhanced Traffic Sign Recognition. (80%)
Projected Randomized Smoothing for Certified Adversarial Robustness. (76%)
Combining Two Adversarial Attacks Against Person Re-Identification Systems. (73%)
Seeing Is Not Always Believing: Invisible Collision Attack and Defence on Pre-Trained Models. (2%)
2023-09-23
Defending Pre-trained Language Models as Few-shot Learners against Backdoor Attacks. (61%)
Moving Target Defense based Secured Network Slicing System in the O-RAN Architecture. (1%)
Detecting and Mitigating System-Level Anomalies of Vision-Based Controllers. (1%)
2023-09-22
RBFormer: Improve Adversarial Robustness of Transformer by Robust Bias. (99%)
Spatial-frequency channels, shape bias, and adversarial robustness. (69%)
VIC-KD: Variance-Invariance-Covariance Knowledge Distillation to Make Keyword Spotting More Robust Against Adversarial Attacks. (69%)
Understanding Deep Gradient Leakage via Inversion Influence Functions. (15%)
Pixel-wise Smoothing for Certified Robustness against Camera Motion Perturbations. (10%)
Privacy Assessment on Reconstructed Images: Are Existing Evaluation Metrics Faithful to Human Perception? (5%)
Expressive variational quantum circuits provide inherent privacy in federated learning. (1%)
On Data Fabrication in Collaborative Vehicular Perception: Attacks and Countermeasures. (1%)
2023-09-21
Improving Machine Learning Robustness via Adversarial Training. (99%)
Goal-Oriented Prompt Attack and Safety Evaluation for LLMs. (69%)
HANS, are you clever? Clever Hans Effect Analysis of Neural Systems. (45%)
On the Relationship between Skill Neurons and Robustness in Prompt Tuning. (12%)
DeepTheft: Stealing DNN Model Architectures through Power Side Channel. (1%)
2023-09-20
How Robust is Google's Bard to Adversarial Image Attacks? (99%)
PRAT: PRofiling Adversarial aTtacks. (99%)
When to Trust AI: Advances and Challenges for Certification of Neural Networks. (64%)
AudioFool: Fast, Universal and synchronization-free Cross-Domain Attack on Speech Recognition. (54%)
Understanding Pose and Appearance Disentanglement in 3D Human Pose Estimation. (54%)
Fed-LSAE: Thwarting Poisoning Attacks against Federated Cyber Threat Detection System via Autoencoder-based Latent Space Inspection. (5%)
Compilation as a Defense: Enhancing DL Model Attack Robustness via Tensor Optimization. (2%)
Generalized Face Forgery Detection via Adaptive Learning for Pre-trained Vision Transformer. (1%)
2023-09-19
Language Guided Adversarial Purification. (99%)
What Learned Representations and Influence Functions Can Tell Us About Adversarial Examples. (99%)
Adversarial Attacks Against Uncertainty Quantification. (99%)
Model Leeching: An Extraction Attack Targeting LLMs. (76%)
Information Leakage from Data Updates in Machine Learning Models. (16%)
Robin: A Novel Method to Produce Robust Interpreters for Deep Learning-Based Code Classifiers. (16%)
SPFL: A Self-purified Federated Learning Method Against Poisoning Attacks. (12%)
It's Simplex! Disaggregating Measures to Improve Certified Robustness. (11%)
Nebula: Self-Attention for Dynamic Malware Analysis. (5%)
Extreme Image Transformations Facilitate Robust Latent Object Representations. (1%)
2023-09-18
Stealthy Physical Masked Face Recognition Attack via Adversarial Style Optimization. (99%)
Transferable Adversarial Attack on Image Tampering Localization. (99%)
Efficient Low-Rank GNN Defense Against Structural Attacks. (96%)
Evaluating Adversarial Robustness with Expected Viable Performance. (45%)
Dual Student Networks for Data-Free Model Stealing. (26%)
Securing Fixed Neural Network Steganography. (5%)
GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts. (4%)
Spoofing attack augmentation: can differently-trained attack models improve generalisation? (3%)
Frame-to-Utterance Convergence: A Spectra-Temporal Approach for Unified Spoofing Detection. (1%)
2023-09-17
Reducing Adversarial Training Cost with Gradient Approximation. (99%)
Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM. (61%)
2023-09-16
Context-aware Adversarial Attack on Named Entity Recognition. (99%)
Inverse classification with logistic and softmax classifiers: efficient optimization. (56%)
Robust Backdoor Attacks on Object Detection in Real World. (11%)
Conditional Mutual Information Constrained Deep Learning for Classification. (5%)
2023-09-15
Adversarial Attacks on Tables with Entity Swap. (92%)
HINT: Healthy Influential-Noise based Training to Defend against Data Poisoning Attacks. (87%)
Distributionally Robust Post-hoc Classifiers under Prior Shifts. (1%)
A Duty to Forget, a Right to be Assured? Exposing Vulnerabilities in Machine Unlearning Services. (1%)
2023-09-14
Unleashing the Adversarial Facet of Software Debloating. (98%)
SLMIA-SR: Speaker-Level Membership Inference Attacks against Speaker Recognition Systems. (76%)
What Matters to Enhance Traffic Rule Compliance of Imitation Learning for Automated Driving. (50%)
BAGEL: Backdoor Attacks against Federated Contrastive Learning. (16%)
Physical Invisible Backdoor Based on Camera Imaging. (2%)
M3Dsynth: A dataset of medical 3D images with AI-generated local manipulations. (1%)
2023-09-13
Semantic Adversarial Attacks via Diffusion Models. (99%)
Hardening RGB-D Object Recognition Systems against Adversarial Patch Attacks. (99%)
Mitigating Adversarial Attacks in Federated Learning with Trusted Execution Environments. (99%)
PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection. (99%)
APICom: Automatic API Completion via Prompt Learning and Adversarial Training-based Data Augmentation. (92%)
RAIN: Your Language Models Can Align Themselves without Finetuning. (83%)
Differentiable JPEG: The Devil is in the Details. (70%)
Deep Nonparametric Convexified Filtering for Computational Photography, Image Synthesis and Adversarial Defense. (41%)
MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems. (38%)
Client-side Gradient Inversion Against Federated Learning from Poisoning. (22%)
Safe Reinforcement Learning with Dual Robustness. (1%)
2023-09-12
Using Reed-Muller Codes for Classification with Rejection and Recovery. (99%)
Certified Robust Models with Slack Control and Large Lipschitz Constants. (98%)
Exploring Non-additive Randomness on ViT against Query-Based Black-Box Attacks. (98%)
Compiled Models, Built-In Exploits: Uncovering Pervasive Bit-Flip Attack Surfaces in DNN Executables. (83%)
Backdoor Attacks and Countermeasures in Natural Language Processing Models: A Comprehensive Security Review. (61%)
CToMP: A Cycle-task-oriented Memory Protection Scheme for Unmanned Systems. (8%)
Language Models as Black-Box Optimizers for Vision-Language Models. (4%)
2023-09-11
Generalized Attacks on Face Verification Systems. (88%)
Adversarial Attacks Assessment of Salient Object Detection via Symbolic Learning. (76%)
Exploiting Machine Unlearning for Backdoor Attacks in Deep Learning System. (68%)
Privacy Side Channels in Machine Learning Systems. (10%)
Divergences in Color Perception between Deep Neural Networks and Humans. (4%)
Catch You Everything Everywhere: Guarding Textual Inversion via Concept Watermarking. (1%)
Optimize Weight Rounding via Signed Gradient Descent for the Quantization of LLMs. (1%)
2023-09-10
Outlier Robust Adversarial Training. (98%)
DAD++: Improved Data-free Test Time Adversarial Defense. (98%)
Machine Translation Models Stand Strong in the Face of Adversarial Attacks. (86%)
Secure Set-Based State Estimation for Linear Systems under Adversarial Attacks on Sensors. (12%)
2023-09-09
Towards Robust Model Watermark via Reducing Parametric Vulnerability. (8%)
RecAD: Towards A Unified Library for Recommender Attack and Defense. (1%)
2023-09-08
Exploring Robust Features for Improving Adversarial Robustness. (99%)
ARRTOC: Adversarially Robust Real-Time Optimization and Control. (2%)
Adversarial attacks on hybrid classical-quantum Deep Learning models for Histopathological Cancer Detection. (1%)
Counterfactual Explanations via Locally-guided Sequential Algorithmic Recourse. (1%)
2023-09-07
Experimental Study of Adversarial Attacks on ML-based xApps in O-RAN. (99%)
How adversarial attacks can disrupt seemingly stable accurate classifiers. (99%)
Adversarially Robust Deep Learning with Optimal-Transport-Regularized Divergences. (95%)
DiffDefense: Defending against Adversarial Attacks via Diffusion Models. (80%)
One-to-Multiple Clean-Label Image Camouflage (OmClic) based Backdoor Attack on Deep Learning. (73%)
Promoting Fairness in GNNs: A Characterization of Stability. (1%)
2023-09-06
Certifying LLM Safety against Adversarial Prompting. (99%)
SWAP: Exploiting Second-Ranked Logits for Adversarial Attacks on Time Series. (84%)
Byzantine-Robust Federated Learning with Variance Reduction and Differential Privacy. (68%)
J-Guard: Journalism Guided Adversarially Robust Detection of AI-generated News. (38%)
MIRA: Cracking Black-box Watermarking on Deep Neural Networks via Model Inversion-based Removal Attacks. (22%)
My Art My Choice: Adversarial Protection Against Unruly AI. (2%)
VeriDIP: Verifying Ownership of Deep Neural Networks through Privacy Leakage Fingerprints. (1%)
A Theoretical Explanation of Activation Sparsity through Flat Minima and Adversarial Robustness. (1%)
2023-09-05
The Adversarial Implications of Variable-Time Inference. (99%)
Adaptive Adversarial Training Does Not Increase Recourse Costs. (92%)
Black-Box Attacks against Signed Graph Analysis via Balance Poisoning. (87%)
RobustEdge: Low Power Adversarial Detection for Cloud-Edge Systems. (83%)
Building a Winning Team: Selecting Source Model Ensembles using a Submodular Transferability Estimation Approach. (4%)
Robust Recommender System: A Survey and Future Directions. (2%)
Dual Adversarial Alignment for Realistic Support-Query Shift Few-shot Learning. (1%)
2023-09-04
Hindering Adversarial Attacks with Multiple Encrypted Patch Embeddings. (99%)
Improving Visual Quality and Transferability of Adversarial Attacks on Face Recognition Simultaneously with Adversarial Restoration. (99%)
Adv3D: Generating 3D Adversarial Examples in Driving Scenarios with NeRF. (99%)