{"total":109,"rows":[{"id":"2410.18927","title":"SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models","created":"2024-10-24","cited_by_count":69,"area":"safety"},{"id":"2502.14881","title":"A Survey of Safety on Large Vision-Language Models: Attacks, Defenses and Evaluations","created":"2025-02-14","cited_by_count":67,"area":"mllm"},{"id":"2502.11054","title":"Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models","created":"2025-02-16","cited_by_count":64,"area":"lr"},{"id":"2506.19676","title":"A Survey of LLM-Driven AI Agent Communication: Protocols, Security Risks, and Defense Countermeasures","created":"2025-06-24","cited_by_count":57,"area":"agents"},{"id":"2409.15310","title":"Visual Prompting in Multimodal Large Language Models: A Survey","created":"2024-09-05","cited_by_count":54,"area":"mllm"},{"id":"2410.07112","title":"VHELM: A Holistic Evaluation of Vision Language Models","created":"2024-10-09","cited_by_count":53,"area":"mllm"},{"id":"2410.05346","title":"AnyAttack: Towards Large-scale Self-supervised Adversarial Attacks on Vision-language Models","created":"2024-10-07","cited_by_count":46,"area":"mllm"},{"id":"2411.18688","title":"Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment","created":"2024-11-27","cited_by_count":44,"area":"lr"},{"id":"2504.14348","title":"Manipulating Multimodal Agents via Cross-Modal Prompt Injection","created":"2025-04-19","cited_by_count":44,"area":"mllm"},{"id":"2406.06302","title":"Unveiling the Safety of GPT-4o: An Empirical Study using Jailbreak Attacks","created":"2024-06-10","cited_by_count":38,"area":"safety"},{"id":"2506.14697","title":"AGENTSAFE: Benchmarking the Safety of Embodied Agents on Hazardous Instructions","created":"2025-06-17","cited_by_count":38,"area":"mllm"},{"id":"2410.23861","title":"Audio Is the Achilles' Heel: Red Teaming Audio Large Multimodal Models","created":"2024-10-31","cited_by_count":37,"area":"mllm"},{"id":"2411.18275","title":"Visual Adversarial Attack on Vision-Language Models for Autonomous Driving","created":"2024-11-27","cited_by_count":36,"area":"lr"},{"id":"2412.05934","title":"Heuristic-Induced Multimodal Risk Distribution Jailbreak Attack for Multimodal Large Language Models","created":"2024-12-08","cited_by_count":33,"area":"safety"},{"id":"2411.09259","title":"Jailbreak Attacks and Defenses against Multimodal Generative Models: A Survey","created":"2024-11-14","cited_by_count":32,"area":"safety"},{"id":"2410.20971","title":"BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks","created":"2024-10-28","cited_by_count":28,"area":"mllm"},{"id":"2503.15092","title":"Towards Understanding the Safety Boundaries of DeepSeek Models: Evaluation and Findings","created":"2025-03-19","cited_by_count":27,"area":null},{"id":"2410.23687","title":"Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey","created":"2024-10-31","cited_by_count":23,"area":"mllm"},{"id":"2505.06679","title":"T2V-OptJail: Discrete Prompt Optimization for Text-to-Video Jailbreak Attacks","created":"2025-05-10","cited_by_count":21,"area":"diffusion"},{"id":"2510.10073","title":"SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents","created":"2025-10-11","cited_by_count":19,"area":"agents"},{"id":"2504.15512","title":"T2VShield: Model-Agnostic Jailbreak Defense for Text-to-Video Models","created":"2025-04-22","cited_by_count":18,"area":"safety"},{"id":"2412.05892","title":"PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximization","created":"2024-12-08","cited_by_count":17,"area":"mllm"},{"id":"2505.21277","title":"Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space","created":"2025-05-27","cited_by_count":17,"area":"safety"},{"id":"2507.00841","title":"SafeMobile: Chain-level Jailbreak Detection and Automated Evaluation for Multimodal Mobile Agents","created":"2025-07-01","cited_by_count":15,"area":"safety"},{"id":"2512.21220","title":"RoboSafe: Safeguarding Embodied Agents via Executable Safety Logic","created":"2025-12-24","cited_by_count":14,"area":"agents"},{"id":"2602.18739","title":"When World Models Dream Wrong: Physical-Conditioned Adversarial Attacks against World Models","created":"2026-02-21","cited_by_count":13,"area":"wm"},{"id":"2411.18000","title":"Exploring Visual Vulnerabilities via Multi-Loss Adversarial Search for Jailbreaking Vision-Language Models","created":"2024-11-27","cited_by_count":12,"area":"mllm"},{"id":"2503.04833","title":"E$^2$AT: Multimodal Jailbreak Defense via Dynamic Joint Optimization for Multimodal Large Language Models","created":"2025-03-05","cited_by_count":12,"area":"mllm"},{"id":"2505.19911","title":"Attention! Your Vision Language Model Could Be Maliciously Manipulated","created":"2025-05-26","cited_by_count":12,"area":"mllm"},{"id":"2506.12430","title":"Pushing the Limits of Safety: A Technical Report on the ATLAS Challenge 2025","created":"2025-06-14","cited_by_count":12,"area":"mllm"},{"id":"2512.06589","title":"OmniSafeBench-MM: A Unified Benchmark and Toolbox for Multimodal Jailbreak Attack-Defense Evaluation","created":"2025-12-06","cited_by_count":11,"area":"safety"},{"id":"2411.11496","title":"Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models","created":"2024-11-18","cited_by_count":10,"area":"lr"},{"id":"2510.01243","title":"Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editing","created":"2025-09-24","cited_by_count":10,"area":"alignment"},{"id":"2506.01307","title":"Align is not Enough: Multimodal Universal Jailbreak Attack against Multimodal Large Language Models","created":"2025-06-02","cited_by_count":9,"area":"mllm"},{"id":"2502.00718","title":"\"I am bad\": Interpreting Stealthy, Universal and Robust Audio Jailbreaks in Audio-Language Models","created":"2025-02-02","cited_by_count":8,"area":"safety"},{"id":"2505.16241","title":"Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers","created":"2025-05-22","cited_by_count":8,"area":"lr"},{"id":"2512.20168","title":"Odysseus: Jailbreaking Commercial Multimodal LLM-integrated Systems via Dual Steganography","created":"2025-12-23","cited_by_count":8,"area":"mllm"},{"id":"2603.13435","title":"CtrlAttack: A Unified Attack on World-Model Control in Diffusion Models","created":"2026-03-13","cited_by_count":7,"area":"wm"},{"id":"2505.06579","title":"POISONCRAFT: Practical Poisoning of Retrieval-Augmented Generation for Large Language Models","created":"2025-05-10","cited_by_count":6,"area":"rag"},{"id":"2509.06350","title":"Mask-GCG: Are All Tokens in Adversarial Suffixes Necessary for Jailbreak Attacks?","created":"2025-09-08","cited_by_count":6,"area":"safety"},{"id":"2604.05498","title":"JailWAM: Jailbreaking World Action Models in Robot Control","created":"2026-04-07","cited_by_count":6,"area":"wm"},{"id":"2505.04673","title":"REVEAL: Multi-turn Evaluation of Image-Input Harms for Vision LLM","created":"2025-05-07","cited_by_count":5,"area":null},{"id":"2510.02677","title":"ARMs: Adaptive Red-Teaming Agent against Multimodal Models with Plug-and-Play Attacks","created":"2025-10-03","cited_by_count":5,"area":"mllm"},{"id":"2410.01438","title":"The Great Contradiction Showdown: How Jailbreak and Stealth Wrestle in Vision-Language Models?","created":"2024-10-02","cited_by_count":4,"area":"mllm"},{"id":"2412.05232","title":"LIAR: Leveraging Inference Time Alignment (Best-of-N) to Jailbreak LLMs in Seconds","created":"2024-12-06","cited_by_count":4,"area":"safety"},{"id":"2503.11750","title":"Making Every Step Effective: Jailbreaking Large Vision-Language Models Through Hierarchical KV Equalization","created":"2025-03-14","cited_by_count":4,"area":"mllm"},{"id":"2505.07258","title":"No Query, No Access","created":"2025-05-12","cited_by_count":4,"area":null},{"id":"2505.16104","title":"Hierarchical Safety Realignment: Lightweight Restoration of Safety in Pruned Large Vision-Language Models","created":"2025-05-22","cited_by_count":4,"area":"mllm"},{"id":"2508.15182","title":"SafeLLM: Unlearning Harmful Outputs from Large Language Models against Jailbreak Attacks","created":"2025-08-21","cited_by_count":4,"area":"safety"},{"id":"2508.19445","title":"On Surjectivity of Neural Networks: Can you elicit any behavior from your model?","created":"2025-08-26","cited_by_count":4,"area":null},{"id":"2601.06460","title":"Tone Matters: The Impact of Linguistic Tone on Hallucination in VLMs","created":"2026-01-10","cited_by_count":4,"area":"mllm"},{"id":"2603.27918","title":"Adversarial Attacks on Multimodal Large Language Models: A Comprehensive Survey","created":"2026-03-30","cited_by_count":4,"area":"mllm"},{"id":"2604.06950","title":"Making MLLMs Blind: Adversarial Smuggling Attacks in MLLM Content Moderation","created":"2026-04-08","cited_by_count":4,"area":"mllm"},{"id":"2506.15734","title":"The Safety Reminder: A Soft Prompt to Reactivate Delayed Safety Awareness in Vision-Language Models","created":"2025-06-15","cited_by_count":3,"area":"mllm"},{"id":"2507.21540","title":"PRISM: Programmatic Reasoning with Image Sequence Manipulation for LVLM Jailbreaking","created":"2025-07-29","cited_by_count":3,"area":"lr"},{"id":"2508.09105","title":"SMA: Who Said That? Auditing Membership Leakage in Semi-Black-box RAG Controlling","created":"2025-08-12","cited_by_count":3,"area":"rag"},{"id":"2602.11096","title":"Safety Recovery in Reasoning Models Is Only a Few Early Steering Steps Away","created":"2026-02-11","cited_by_count":3,"area":"lr"},{"id":"2604.18803","title":"LLM-as-Judge Framework for Evaluating Tone-Induced Hallucination in Vision-Language Models","created":"2026-04-20","cited_by_count":3,"area":"mllm"},{"id":"2503.06989","title":"Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application","created":"2025-03-10","cited_by_count":2,"area":"mllm"},{"id":"2503.11619","title":"Tit-for-Tat: Safeguarding Large Vision-Language Models Against Jailbreak Attacks via Adversarial Defense","created":"2025-03-14","cited_by_count":2,"area":"safety"},{"id":"2506.03683","title":"PRJ: Perception-Retrieval-Judgement for Generated Images","created":"2025-06-04","cited_by_count":2,"area":null},{"id":"2508.10955","title":"Empowering Multimodal LLMs with External Tools: A Comprehensive Survey","created":"2025-08-14","cited_by_count":2,"area":"mllm"},{"id":"2509.21401","title":"JaiLIP: Jailbreaking Vision-Language Models via Loss Guided Image Perturbation","created":"2025-09-24","cited_by_count":2,"area":"lr"},{"id":"2510.15068","title":"Sequential Comics for Jailbreaking Multimodal Large Language Models via Structured Visual Storytelling","created":"2025-10-16","cited_by_count":2,"area":"mllm"},{"id":"2601.22398","title":"Jailbreaks on Vision Language Model via Multimodal Reasoning","created":"2026-01-29","cited_by_count":2,"area":"lr"},{"id":"2602.10179","title":"When the Prompt Becomes Visual: Vision-Centric Jailbreak Attacks for Large Image Editing Models","created":"2026-02-10","cited_by_count":2,"area":"safety"},{"id":"2603.05872","title":"Evolving Deception: When Agents Evolve, Deception Wins","created":"2026-03-06","cited_by_count":2,"area":"agents"},{"id":"2603.27332","title":"Unsafe by Reciprocity: How Generation-Understanding Coupling Undermines Safety in Unified Multimodal Models","created":"2026-03-28","cited_by_count":2,"area":null},{"id":"2606.31227","title":"Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming","created":"2026-06-30","cited_by_count":2,"area":"agents"},{"id":"2412.16555","title":"Divide and Conquer: A Hybrid Strategy Defeats Multimodal Large Language Models","created":"2024-12-21","cited_by_count":1,"area":"mllm"},{"id":"2505.21556","title":"Benign-to-Toxic Jailbreaking: Inducing Harmful Responses from Harmless Prompts","created":"2025-05-26","cited_by_count":1,"area":"mllm"},{"id":"2507.00052","title":"VSF-Med:A Vulnerability Scoring Framework for Medical Vision-Language Models","created":"2025-06-25","cited_by_count":1,"area":"mllm"},{"id":"2508.10404","title":"Layer-Wise Perturbations via Sparse Autoencoders for Adversarial Text Generation","created":"2025-08-14","cited_by_count":1,"area":"safety"},{"id":"2508.16406","title":"Retrieval-Augmented Defense: Adaptive and Controllable Jailbreak Prevention for Large Language Models","created":"2025-08-22","cited_by_count":1,"area":"rag"},{"id":"2510.09699","title":"VisualDAN: Exposing Vulnerabilities in VLMs with Visual-Driven DAN Commands","created":"2025-10-09","cited_by_count":1,"area":"mllm"},{"id":"2512.16921","title":"Differences That Matter: Auditing Models for Capability Gap Discovery and Rectification","created":"2025-12-18","cited_by_count":1,"area":"mllm"},{"id":"2602.08136","title":"Robustness of Vision Language Models Against Split-Image Harmful Input Attacks","created":"2026-02-08","cited_by_count":1,"area":"mllm"},{"id":"2603.05921","title":"BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviation","created":"2026-03-06","cited_by_count":1,"area":"safety"},{"id":"2603.14825","title":"Two Birds, One Projection: Harmonizing Safety and Utility in LVLMs via Inference-time Feature Projection","created":"2026-03-16","cited_by_count":1,"area":"mllm"},{"id":"2604.04630","title":"Multimodal Backdoor Attack on VLMs for Autonomous Driving via Graffiti and Cross-Lingual Triggers","created":"2026-04-06","cited_by_count":1,"area":"mllm"},{"id":"2605.18915","title":"DMN: A Compositional Framework for Jailbreaking Multimodal LLMs with Multi-Image Inputs","created":"2026-05-18","cited_by_count":1,"area":"mllm"},{"id":"2606.25592","title":"VPA-Guard: Defending and Benchmarking Image-to-Video Generation Against Visual Prompt Attacks","created":"2026-06-24","cited_by_count":1,"area":null},{"id":"2606.26566","title":"Adversarial Diffusion Across Modalities: A Fusion Survey of Attacks, Defenses, and Evaluation for Text, Vision, and Vision-Language Models","created":"2026-06-25","cited_by_count":1,"area":"mllm"},{"id":"2607.05783","title":"Benchmarking the Robustness of Autonomous Driving to Environmental Illusions: A Lane Perception Perspective","created":"2026-07-07","cited_by_count":1,"area":"driving"},{"id":"2608.01117","title":"SoK: Intent-Oriented Systematization of Multi-Turn LLM Jailbreaks","created":"2026-08-02","cited_by_count":1,"area":"safety"},{"id":"2507.20994","title":"Security Tensors as a Cross-Modal Bridge: Extending Text-Aligned Safety to Vision in LVLM","created":"2025-07-28","cited_by_count":0,"area":"mllm"},{"id":"2509.12724","title":"Defense-to-Attack: Bypassing Weak Defenses Enables Stronger Jailbreaks in Vision-Language Models","created":"2025-09-16","cited_by_count":0,"area":"safety"},{"id":"2509.21400","title":"SafeSteer: Adaptive Subspace Steering for Efficient Jailbreak Defense in Vision-Language Models","created":"2025-09-24","cited_by_count":0,"area":"safety"},{"id":"2510.20129","title":"SAID: Safety-Aware Intent Defense via Prefix Probing for Large Language Models","created":"2025-10-23","cited_by_count":0,"area":"alignment"},{"id":"2510.21214","title":"Enhanced MLLM Black-Box Jailbreaking Attacks and Defenses","created":"2025-10-24","cited_by_count":0,"area":"safety"},{"id":"2511.20002","title":"Semantic Router: On the Feasibility of Hijacking MLLMs via a Single Adversarial Perturbation","created":"2025-11-25","cited_by_count":0,"area":"mllm"},{"id":"2603.09246","title":"Reasoning-Oriented Programming: Chaining Semantic Gadgets to Jailbreak Large Vision Language Models","created":"2026-03-10","cited_by_count":0,"area":"alignment"},{"id":"2603.12799","title":"What Makes VLMs Robust? Towards Reconciling Robustness and Accuracy in Vision-Language Models","created":"2026-03-13","cited_by_count":0,"area":"mllm"},{"id":"2604.04488","title":"A Patch-based Cross-view Regularized Framework for Backdoor Defense in Multimodal Large Language Models","created":"2026-04-06","cited_by_count":0,"area":"mllm"},{"id":"2604.05853","title":"Reading Between the Pixels: An Inscriptive Jailbreak Attack on Text-to-Image Models","created":"2026-04-07","cited_by_count":0,"area":"safety"},{"id":"2604.10299","title":"Seeing No Evil: Blinding Large Vision-Language Models to Safety Instructions via Adversarial Attention Hijacking","created":"2026-04-11","cited_by_count":0,"area":"mllm"},{"id":"2605.22273","title":"Exposing Vulnerabilities in Visible-Infrared VLMs: A Unified Geometric Adversarial Framework with Cross-Task Transferability","created":"2026-05-21","cited_by_count":0,"area":"mllm"},{"id":"2605.25194","title":"Localize and Neutralize: Gradient-guided Token Suppression against Visual Prompt Injection Attack","created":"2026-05-24","cited_by_count":0,"area":"safety"},{"id":"2605.25534","title":"StructBreak: Structural Cognitive Overload-Induced Safety Failures in MLLMs","created":"2026-05-25","cited_by_count":0,"area":"lr"},{"id":"2605.27927","title":"Structure-Guided Visual Perturbation Neutralization for LVLMs","created":"2026-05-27","cited_by_count":0,"area":"mllm"}],"next_cursor":"100"}