{"total":128,"rows":[{"id":"2404.02151","title":"Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks","created":"2024-04-02","cited_by_count":486,"area":"safety"},{"id":"2312.02003","title":"A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Ugly","created":"2023-12-04","cited_by_count":387,"area":null},{"id":"2402.08787","title":"Rethinking Machine Unlearning for Large Language Models","created":"2024-02-13","cited_by_count":257,"area":"lr"},{"id":"2402.08679","title":"COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability","created":"2024-02-13","cited_by_count":175,"area":"safety"},{"id":"2407.15549","title":"Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs","created":"2024-07-22","cited_by_count":173,"area":"safety"},{"id":"2401.14446","title":"Black-Box Access is Insufficient for Rigorous AI Audits","created":"2024-01-25","cited_by_count":126,"area":null},{"id":"2504.15585","title":"A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment","created":"2025-04-22","cited_by_count":116,"area":"lr"},{"id":"2401.17256","title":"Weak-to-Strong Jailbreaking on Large Language Models","created":"2024-01-30","cited_by_count":107,"area":"safety"},{"id":"2402.09063","title":"Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space","created":"2024-02-14","cited_by_count":91,"area":"alignment"},{"id":"2401.15897","title":"Red-Teaming for Generative AI: Silver Bullet or Security Theater?","created":"2024-01-29","cited_by_count":83,"area":"safety"},{"id":"2410.02832","title":"FlipAttack: Jailbreak LLMs via Flipping","created":"2024-10-02","cited_by_count":80,"area":"safety"},{"id":"2510.07192","title":"Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples","created":"2025-10-08","cited_by_count":72,"area":"safety"},{"id":"2405.13401","title":"TrojanRAG: Retrieval-Augmented Generation Can Be Backdoor Driver in Large Language Models","created":"2024-05-22","cited_by_count":67,"area":"rag"},{"id":"2406.17092","title":"BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models","created":"2024-06-24","cited_by_count":58,"area":"safety"},{"id":"2504.20984","title":"ACE: A Security Architecture for LLM-Integrated App Systems","created":"2025-04-29","cited_by_count":55,"area":null},{"id":"2408.12798","title":"BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models","created":"2024-08-23","cited_by_count":50,"area":"safety"},{"id":"2405.20485","title":"Phantom: General Backdoor Attacks on Retrieval Augmented Language Generation","created":"2024-05-30","cited_by_count":49,"area":"rag"},{"id":"2410.13722","title":"Persistent Pre-Training Poisoning of LLMs","created":"2024-10-17","cited_by_count":48,"area":null},{"id":"2403.01251","title":"Accelerating Greedy Coordinate Gradient and General Prompt Optimization via Probe Sampling","created":"2024-03-02","cited_by_count":37,"area":"safety"},{"id":"2402.16382","title":"Immunization against harmful fine-tuning attacks","created":"2024-02-26","cited_by_count":35,"area":null},{"id":"2404.05530","title":"Best-of-Venom: Attacking RLHF by Injecting Poisoned Preference Data","created":"2024-04-08","cited_by_count":35,"area":"alignment"},{"id":"2410.08811","title":"PoisonBench: Assessing Large Language Model Vulnerability to Data Poisoning","created":"2024-10-11","cited_by_count":33,"area":"safety"},{"id":"2402.14968","title":"Mitigating Fine-tuning based Jailbreak Attack with Backdoor Enhanced Safety Alignment","created":"2024-02-22","cited_by_count":31,"area":null},{"id":"2404.14461","title":"Competition Report: Finding Universal Jailbreak Backdoors in Aligned LLMs","created":"2024-04-22","cited_by_count":28,"area":"safety"},{"id":"2402.15302","title":"How (un)ethical are instruction-centric responses of LLMs? Unveiling the vulnerabilities of safety guardrails to harmful queries","created":"2024-02-23","cited_by_count":27,"area":"safety"},{"id":"2409.14457","title":"Large Model Based Agents: State-of-the-Art, Cooperation Paradigms, Security and Privacy, and Future Trends","created":"2024-09-22","cited_by_count":26,"area":null},{"id":"2504.13146","title":"Antidistillation Sampling","created":"2025-04-17","cited_by_count":25,"area":null},{"id":"2406.14859","title":"From LLMs to MLLMs: Exploring the Landscape of Multimodal Jailbreaking","created":"2024-06-21","cited_by_count":24,"area":"mllm"},{"id":"2409.11564","title":"Preference Tuning with Human Feedback on Language, Speech, and Vision Tasks: A Survey","created":"2024-09-17","cited_by_count":23,"area":"alignment"},{"id":"2410.14827","title":"Enhancing Prompt Injection Attacks to LLMs via Poisoning Alignment","created":"2024-10-18","cited_by_count":21,"area":"safety"},{"id":"2409.19998","title":"Do Influence Functions Work on Large Language Models?","created":"2024-09-30","cited_by_count":20,"area":null},{"id":"2503.22759","title":"Data Poisoning in Deep Learning: A Survey","created":"2025-03-27","cited_by_count":20,"area":"safety"},{"id":"2506.16447","title":"Probe before You Talk: Towards Black-box Defense against Backdoor Unalignment for Large Language Models","created":"2025-06-19","cited_by_count":19,"area":"alignment"},{"id":"2602.11865","title":"Intelligent AI Delegation","created":"2026-02-12","cited_by_count":18,"area":null},{"id":"2405.19524","title":"AI Risk Management Should Incorporate Both Safety and Security","created":"2024-05-29","cited_by_count":17,"area":null},{"id":"2501.17858","title":"Improving Your Model Ranking on Chatbot Arena by Vote Rigging","created":"2025-01-29","cited_by_count":17,"area":null},{"id":"2402.13459","title":"Learning to Poison Large Language Models for Downstream Manipulation","created":"2024-02-21","cited_by_count":16,"area":"lr"},{"id":"2501.17186","title":"Complete Chess Games Enable LLM Become A Chess Master","created":"2025-01-26","cited_by_count":16,"area":null},{"id":"2407.04108","title":"Future Events as Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs","created":"2024-07-04","cited_by_count":15,"area":"safety"},{"id":"2409.19993","title":"Mitigating Backdoor Threats to Large Language Models: Advancement and Challenges","created":"2024-09-30","cited_by_count":15,"area":"safety"},{"id":"2502.02260","title":"Position: Adversarial ML for LLMs Is Not Making Any Progress","created":"2025-02-04","cited_by_count":15,"area":null},{"id":"2402.01920","title":"Preference Poisoning Attacks on Reward Model Learning","created":"2024-02-02","cited_by_count":14,"area":"alignment"},{"id":"2504.18598","title":"BadMoE: Backdooring Mixture-of-Experts LLMs via Optimizing Routing Triggers and Infecting Dormant Experts","created":"2025-04-24","cited_by_count":14,"area":"moe"},{"id":"2505.17767","title":"Position: The Real Barrier to LLM Agent Usability is Agentic ROI","created":"2025-05-23","cited_by_count":12,"area":"agents"},{"id":"2406.14393","title":"Jailbreaking as a Reward Misspecification Problem","created":"2024-06-20","cited_by_count":11,"area":"alignment"},{"id":"2409.00787","title":"The Dark Side of Human Feedback: Poisoning Large Language Models via User Inputs","created":"2024-09-01","cited_by_count":10,"area":"alignment"},{"id":"2505.17601","title":"Revisiting Backdoor Attacks on LLMs: A Stealthy and Practical Poisoning Framework via Harmless Inputs","created":"2025-05-23","cited_by_count":10,"area":"safety"},{"id":"2509.21761","title":"Backdoor Attribution: Elucidating and Controlling Backdoor in Language Models","created":"2025-09-26","cited_by_count":10,"area":"safety"},{"id":"2409.00399","title":"Rethinking Backdoor Detection Evaluation for Language Models","created":"2024-08-31","cited_by_count":9,"area":"lr"},{"id":"2505.24842","title":"Cascading Adversarial Bias from Injection to Distillation in Language Models","created":"2025-05-30","cited_by_count":9,"area":null},{"id":"2507.01321","title":"ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks","created":"2025-07-02","cited_by_count":9,"area":"safety"},{"id":"2506.06518","title":"A Systematic Review of Poisoning Attacks Against Large Language Models","created":"2025-06-06","cited_by_count":8,"area":null},{"id":"2508.20697","title":"Token Buncher: Shielding LLMs from Harmful Reinforcement Learning Fine-Tuning","created":"2025-08-28","cited_by_count":8,"area":null},{"id":"2601.04034","title":"HoneyTrap: Deceiving Large Language Model Attackers to Honeypot Traps with Resilient Multi-Agent Defense","created":"2026-01-07","cited_by_count":8,"area":"agents"},{"id":"2508.00161","title":"Watch the Weights: Unsupervised monitoring and control of fine-tuned LLMs","created":"2025-07-31","cited_by_count":7,"area":"safety"},{"id":"2510.07985","title":"Fewer Weights, More Problems: A Practical Attack on LLM Pruning","created":"2025-10-09","cited_by_count":7,"area":null},{"id":"2602.06258","title":"GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt","created":"2026-02-05","cited_by_count":7,"area":"alignment"},{"id":"2603.13461","title":"Purifying Generative LLMs from Backdoors without Prior Knowledge or Clean Reference","created":"2026-03-13","cited_by_count":7,"area":"safety"},{"id":"2409.03274","title":"Recent Advances in Attack and Defense Approaches of Large Language Models","created":"2024-09-05","cited_by_count":6,"area":null},{"id":"2505.14300","title":"Beyond Black-Box Obfuscation: Mechanistic Analysis and Defense of White-Box Monitors","created":"2025-05-20","cited_by_count":6,"area":null},{"id":"2510.10265","title":"Backdoor Collapse: Eliminating Unknown Threats via Known Backdoor Aggregation in Language Models","created":"2025-10-11","cited_by_count":6,"area":"safety"},{"id":"2510.17021","title":"Forgetting to Forget: Attention Sink as A Gateway for Backdooring LLM Unlearning","created":"2025-10-19","cited_by_count":6,"area":"safety"},{"id":"2510.18541","title":"Pay Attention to the Triggers: Constructing Backdoors That Survive Distillation","created":"2025-10-21","cited_by_count":6,"area":"distill"},{"id":"2404.16020","title":"Investigating Adversarial Trigger Transfer in Large Language Models","created":"2024-04-24","cited_by_count":5,"area":"alignment"},{"id":"2410.11283","title":"AdvBDGen: Adversarially Fortified Prompt-Specific Fuzzy Backdoor Generator Against LLM Alignment","created":"2024-10-15","cited_by_count":5,"area":"alignment"},{"id":"2503.03039","title":"LLM Misalignment via Adversarial RLHF Platforms","created":"2025-03-04","cited_by_count":5,"area":"alignment"},{"id":"2505.16567","title":"Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM Finetuning","created":"2025-05-22","cited_by_count":5,"area":"safety"},{"id":"2603.27517","title":"A Security Analysis of the OpenClaw AI Agent Framework","created":"2026-03-29","cited_by_count":5,"area":"lr"},{"id":"2311.17228","title":"Survey on AI Ethics: A Socio-technical Perspective","created":"2023-11-28","cited_by_count":4,"area":null},{"id":"2402.12617","title":"Generative AI Security: Challenges and Countermeasures","created":"2024-02-20","cited_by_count":4,"area":null},{"id":"2412.08201","title":"Model-Editing-Based Jailbreak against Safety-aligned Large Language Models","created":"2024-12-11","cited_by_count":4,"area":"alignment"},{"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":"2510.02334","title":"Where Did It Go Wrong? Attributing Undesirable LLM Behaviors via Representation Gradient Tracing","created":"2025-09-26","cited_by_count":4,"area":"safety"},{"id":"2406.05660","title":"Injecting Undetectable Backdoors in Obfuscated Neural Networks and Language Models","created":"2024-06-09","cited_by_count":3,"area":"safety"},{"id":"2511.11020","title":"Data Poisoning Vulnerabilities Across Healthcare AI Architectures: A Security Threat Analysis","created":"2025-11-14","cited_by_count":3,"area":"safety"},{"id":"2604.09748","title":"Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward","created":"2026-04-10","cited_by_count":3,"area":"lr"},{"id":"2605.23168","title":"PoisonForge: Task-Level Targeted Poisoning Benchmark for Instruction-Tuned LLMs","created":"2026-05-22","cited_by_count":3,"area":null},{"id":"2411.07691","title":"New Emerged Security and Privacy of Pre-trained Model: a Survey and Outlook","created":"2024-11-12","cited_by_count":2,"area":null},{"id":"2503.10228","title":"Policy Teaching via Data Poisoning in Learning from Human Preferences","created":"2025-03-13","cited_by_count":2,"area":"alignment"},{"id":"2503.12301","title":"One Goal, Many Challenges: Robust Preference Optimization Amid Content-Aware and Multi-Source Noise","created":"2025-03-16","cited_by_count":2,"area":"alignment"},{"id":"2506.03234","title":"BadReward: Clean-Label Poisoning of Reward Models in Text-to-Image RLHF","created":"2025-06-03","cited_by_count":2,"area":"alignment"},{"id":"2506.11402","title":"LoRA Users Beware: A Few Spurious Tokens Can Manipulate Your Finetuned Model","created":"2025-06-13","cited_by_count":2,"area":null},{"id":"2507.02850","title":"LLM Hypnosis: Exploiting User Feedback for Unauthorized Knowledge Injection to All Users","created":"2025-07-03","cited_by_count":2,"area":"alignment"},{"id":"2507.08020","title":"Circumventing Safety Alignment in Large Language Models Through Embedding Space Toxicity Attenuation","created":"2025-07-08","cited_by_count":2,"area":"alignment"},{"id":"2510.05169","title":"From Poisoned to Aware: Fostering Backdoor Self-Awareness in LLMs","created":"2025-10-05","cited_by_count":2,"area":"safety"},{"id":"2602.22246","title":"Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models","created":"2026-02-24","cited_by_count":2,"area":"diffusion"},{"id":"2606.02995","title":"Patcher: Post-Hoc Patching of Backdoored Large Language Models","created":"2026-06-02","cited_by_count":2,"area":"alignment"},{"id":"2312.14712","title":"Balancing Privacy, Robustness, and Efficiency in Machine Learning","created":"2023-12-22","cited_by_count":1,"area":null},{"id":"2501.07959","title":"Self-Instruct Few-Shot Jailbreaking: Decompose the Attack into Pattern and Behavior Learning","created":"2025-01-14","cited_by_count":1,"area":"synth"},{"id":"2506.13285","title":"DualEdit: Mitigating Safety Fallback in LLM Backdoor Editing via Affirmation-Refusal Regulation","created":"2025-06-16","cited_by_count":1,"area":"alignment"},{"id":"2507.01752","title":"Tuning without Peeking: Provable Generalization Bounds and Robust LLM Post-Training","created":"2025-07-02","cited_by_count":1,"area":"safety"},{"id":"2509.19775","title":"bi-GRPO: Bidirectional Optimization for Jailbreak Backdoor Injection on LLMs","created":"2025-09-24","cited_by_count":1,"area":"lr"},{"id":"2510.09260","title":"GREAT: Generalizable Backdoor Attacks in RLHF via Emotion-Aware Trigger Synthesis","created":"2025-10-10","cited_by_count":1,"area":"alignment"},{"id":"2510.09849","title":"Text Prompt Injection of Vision Language Models","created":"2025-10-10","cited_by_count":1,"area":"mllm"},{"id":"2601.22818","title":"Hide and Seek in Embedding Space: Geometry-based Steganography and Detection in Large Language Models","created":"2026-01-30","cited_by_count":1,"area":"mechinterp"},{"id":"2602.05401","title":"BadTemplate: A Training-Free Backdoor Attack via Chat Template Against Large Language Models","created":"2026-02-05","cited_by_count":1,"area":"safety"},{"id":"2602.09757","title":"Towards Poisoning Robustness Certification for Natural Language Generation","created":"2026-02-10","cited_by_count":1,"area":null},{"id":"2605.02495","title":"Efficient Preference Poisoning Attack on Offline RLHF","created":"2026-05-04","cited_by_count":1,"area":"alignment"},{"id":"2605.16339","title":"Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders","created":"2026-05-07","cited_by_count":1,"area":"alignment"},{"id":"2605.25073","title":"Security in the Fine-Tuning Lifecycle of Large Language Models: Threats, Defenses,Evaluation, and Future Directions","created":"2026-05-24","cited_by_count":1,"area":"safety"}],"next_cursor":"100"}