{"total":110,"rows":[{"id":"2307.15217","title":"Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback","created":"2023-07-27","cited_by_count":759,"area":"alignment"},{"id":"2310.12815","title":"Formalizing and Benchmarking Prompt Injection Attacks and Defenses","created":"2023-10-19","cited_by_count":364,"area":"safety"},{"id":"2306.12001","title":"An Overview of Catastrophic AI Risks","created":"2023-06-21","cited_by_count":222,"area":null},{"id":"2307.16888","title":"Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection","created":"2023-07-31","cited_by_count":157,"area":"safety"},{"id":"2402.11208","title":"Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents","created":"2024-02-17","cited_by_count":137,"area":"agents"},{"id":"2306.17194","title":"On the Exploitability of Instruction Tuning","created":"2023-06-28","cited_by_count":116,"area":"safety"},{"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":"2402.13851","title":"VL-Trojan: Multimodal Instruction Backdoor Attacks against Autoregressive Visual Language Models","created":"2024-02-21","cited_by_count":87,"area":"mllm"},{"id":"2308.12833","title":"Use of LLMs for Illicit Purposes: Threats, Prevention Measures, and Vulnerabilities","created":"2023-08-24","cited_by_count":82,"area":null},{"id":"2401.12242","title":"BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models","created":"2024-01-20","cited_by_count":80,"area":"lr"},{"id":"2310.07676","title":"Composite Backdoor Attacks Against Large Language Models","created":"2023-10-11","cited_by_count":73,"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":"2310.10383","title":"Privacy in Large Language Models: Attacks, Defenses and Future Directions","created":"2023-10-16","cited_by_count":51,"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":"2311.09948","title":"Hijacking Large Language Models via Adversarial In-Context Learning","created":"2023-11-16","cited_by_count":45,"area":"safety"},{"id":"2308.14367","title":"A Comprehensive Overview of Backdoor Attacks in Large Language Models within Communication Networks","created":"2023-08-28","cited_by_count":42,"area":"safety"},{"id":"2309.06055","title":"Backdoor Attacks and Countermeasures in Natural Language Processing Models: A Comprehensive Security Review","created":"2023-09-12","cited_by_count":39,"area":"safety"},{"id":"2406.06852","title":"A Survey of Recent Backdoor Attacks and Defenses in Large Language Models","created":"2024-06-10","cited_by_count":36,"area":"safety"},{"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":"2411.12768","title":"CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization","created":"2024-11-18","cited_by_count":33,"area":"safety"},{"id":"2405.17067","title":"Tokenization Matters! Degrading Large Language Models through Challenging Their Tokenization","created":"2024-05-27","cited_by_count":32,"area":null},{"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":"2406.05948","title":"Chain-of-Scrutiny: Detecting Backdoor Attacks for Large Language Models","created":"2024-06-10","cited_by_count":28,"area":"safety"},{"id":"2604.09378","title":"BadSkill: Backdoor Attacks on Agent Skills via Model-in-Skill Poisoning","created":"2026-04-10","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":"2407.20224","title":"Can Editing LLMs Inject Harm?","created":"2024-07-29","cited_by_count":22,"area":"safety"},{"id":"2406.16565","title":"Noisy Neighbors: Efficient membership inference attacks against LLMs","created":"2024-06-24","cited_by_count":21,"area":"safety"},{"id":"2410.14827","title":"Enhancing Prompt Injection Attacks to LLMs via Poisoning Alignment","created":"2024-10-18","cited_by_count":21,"area":"safety"},{"id":"2503.22759","title":"Data Poisoning in Deep Learning: A Survey","created":"2025-03-27","cited_by_count":20,"area":"safety"},{"id":"2503.04636","title":"Mark Your LLM: Detecting the Misuse of Open-Source Large Language Models via Watermarking","created":"2025-03-06","cited_by_count":19,"area":"safety"},{"id":"2602.04863","title":"Subliminal Effects in Your Data: A General Mechanism via Log-Linearity","created":"2026-02-04","cited_by_count":19,"area":null},{"id":"2404.16841","title":"Machine Unlearning in Large Language Models","created":"2024-02-03","cited_by_count":17,"area":"safety"},{"id":"2405.19524","title":"AI Risk Management Should Incorporate Both Safety and Security","created":"2024-05-29","cited_by_count":17,"area":null},{"id":"2409.14729","title":"PROMPTFUZZ: Harnessing Fuzzing Techniques for Robust Testing of Prompt Injection in LLMs","created":"2024-09-23","cited_by_count":17,"area":"safety"},{"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":"2402.12168","title":"Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning","created":"2024-02-19","cited_by_count":16,"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":"2307.16680","title":"On the Trustworthiness Landscape of State-of-the-art Generative Models: A Survey and Outlook","created":"2023-07-31","cited_by_count":15,"area":"diffusion"},{"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":"2506.07214","title":"Backdoor Attack on Vision Language Models with Stealthy Semantic Manipulation","created":"2025-06-08","cited_by_count":14,"area":"mllm"},{"id":"2405.05466","title":"Poser: Unmasking Alignment Faking LLMs by Manipulating Their Internals","created":"2024-05-08","cited_by_count":13,"area":null},{"id":"2309.17319","title":"Building Privacy-Preserving and Secure Geospatial Artificial Intelligence Foundation Models","created":"2023-09-29","cited_by_count":12,"area":null},{"id":"2406.12259","title":"Adversarial Attacks on Large Language Models in Medicine","created":"2024-06-18","cited_by_count":10,"area":null},{"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":"2312.04748","title":"Forcing Generative Models to Degenerate Ones: The Power of Data Poisoning Attacks","created":"2023-12-07","cited_by_count":9,"area":"safety"},{"id":"2403.13309","title":"Mapping LLM Security Landscapes: A Comprehensive Stakeholder Risk Assessment Proposal","created":"2024-03-20","cited_by_count":9,"area":null},{"id":"2407.12281","title":"Turning Generative Models Degenerate: The Power of Data Poisoning Attacks","created":"2024-07-17","cited_by_count":9,"area":"safety"},{"id":"2507.01321","title":"ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks","created":"2025-07-02","cited_by_count":9,"area":"safety"},{"id":"2504.03759","title":"Emerging Cyber Attack Risks of Medical AI Agents","created":"2025-04-02","cited_by_count":8,"area":"agents"},{"id":"2308.11521","title":"Self-Deception: Reverse Penetrating the Semantic Firewall of Large Language Models","created":"2023-08-16","cited_by_count":7,"area":"safety"},{"id":"2510.18333","title":"Position: LLM Watermarking Should Align Stakeholders' Incentives for Practical Adoption","created":"2025-10-21","cited_by_count":7,"area":"safety"},{"id":"2409.01247","title":"Conversational Complexity for Assessing Risk in Large Language Models","created":"2024-09-02","cited_by_count":6,"area":"safety"},{"id":"2504.11182","title":"Exploring Backdoor Attack and Defense for LLM-empowered Recommendations","created":"2025-04-15","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":"2405.14061","title":"Meanings and Feelings of Large Language Models: Observability of Latent States in Generative AI","created":"2024-05-22","cited_by_count":5,"area":null},{"id":"2405.16783","title":"TrojFM: Resource-efficient Backdoor Attacks against Very Large Foundation Models","created":"2024-05-27","cited_by_count":5,"area":"safety"},{"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":"2411.18280","title":"Neutralizing Backdoors through Information Conflicts for Large Language Models","created":"2024-11-27","cited_by_count":5,"area":"safety"},{"id":"2505.15656","title":"Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!","created":"2025-05-21","cited_by_count":5,"area":"safety"},{"id":"2509.07287","title":"Paladin: Defending LLM-enabled Phishing Emails with a New Trigger-Tag Paradigm","created":"2025-09-08","cited_by_count":5,"area":null},{"id":"2507.06850","title":"The Dark Side of LLMs: Agent-based Attack Vectors for System-level Compromise","created":"2025-07-09","cited_by_count":4,"area":"agents"},{"id":"2401.10375","title":"Foundation Models in Federated Learning: Assessing Backdoor Vulnerabilities","created":"2024-01-18","cited_by_count":3,"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":"2410.17573","title":"Securing Federated Learning against Backdoor Threats with Foundation Model Integration","created":"2024-10-23","cited_by_count":3,"area":"safety"},{"id":"2512.19297","title":"Causal-Guided Detoxify Backdoor Attack of Open-Weight LoRA Models","created":"2025-12-22","cited_by_count":3,"area":null},{"id":"2602.04195","title":"Semantic Consensus Decoding: Backdoor Defense for Verilog Code Generation","created":"2026-02-04","cited_by_count":3,"area":"safety"},{"id":"2602.16984","title":"Fundamental Limits of Black-Box Safety Evaluation: Information-Theoretic and Computational Barriers from Latent Context Conditioning","created":"2026-02-19","cited_by_count":3,"area":null},{"id":"2603.23791","title":"The Cognitive Firewall:Securing Browser Based AI Agents Against Indirect Prompt Injection Via Hybrid Edge Cloud Defense","created":"2026-03-24","cited_by_count":3,"area":"agents"},{"id":"2410.23472","title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","created":"2024-10-30","cited_by_count":2,"area":null},{"id":"2504.09026","title":"Detecting Instruction Fine-tuning Attacks using Influence Function","created":"2025-04-12","cited_by_count":2,"area":null},{"id":"2602.22246","title":"Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models","created":"2026-02-24","cited_by_count":2,"area":"diffusion"},{"id":"2604.09235","title":"Unreal Thinking: Chain-of-Thought Hijacking via Two-stage Backdoor","created":"2026-04-10","cited_by_count":2,"area":"safety"},{"id":"2604.17041","title":"SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models","created":"2026-04-18","cited_by_count":2,"area":"mllm"},{"id":"2308.10201","title":"Hiding Backdoors within Event Sequence Data via Poisoning Attacks","created":"2023-08-20","cited_by_count":1,"area":null},{"id":"2405.20234","title":"Hidden in Plain Sight: Exploring Chat History Tampering in Interactive Language Models","created":"2024-05-30","cited_by_count":1,"area":null},{"id":"2408.13221","title":"Protecting against simultaneous data poisoning attacks","created":"2024-08-23","cited_by_count":1,"area":"safety"},{"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":"2503.17578","title":"Large Language Models Can Verbatim Reproduce Long Malicious Sequences","created":"2025-03-21","cited_by_count":1,"area":"safety"},{"id":"2505.23223","title":"Daunce: Data Attribution through Uncertainty Estimation","created":"2025-05-29","cited_by_count":1,"area":null},{"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":"2508.08559","title":"Multi-Target Backdoor Attacks Against Speaker Recognition","created":"2025-08-12","cited_by_count":1,"area":null},{"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":"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":"2603.02262","title":"Silent Sabotage During Fine-Tuning: Few-Shot Rationale Poisoning of Compact Medical LLMs","created":"2026-02-28","cited_by_count":1,"area":"safety"},{"id":"2604.21700","title":"Stealthy Backdoor Attacks against LLMs Based on Natural Style Triggers","created":"2026-04-23","cited_by_count":1,"area":"safety"},{"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"},{"id":"2605.27631","title":"Poison with Style: A Practical Poisoning Attack on Code Large Language Models","created":"2026-05-26","cited_by_count":1,"area":null},{"id":"2509.25525","title":"Defeating Cerberus: Concept-Guided Privacy-Leakage Mitigation in Multimodal Language Models","created":"2025-09-29","cited_by_count":0,"area":"mllm"},{"id":"2510.01288","title":"Microsaccade-Inspired Probing: Positional Encoding Perturbations Reveal LLM Misbehaviours","created":"2025-10-01","cited_by_count":0,"area":null},{"id":"2511.02600","title":"On The Dangers of Poisoned LLMs In Security Automation","created":"2025-11-04","cited_by_count":0,"area":null},{"id":"2602.06887","title":"Plato's Form: Toward Backdoor Defense-as-a-Service for LLMs with Prototype Representations","created":"2026-02-06","cited_by_count":0,"area":"safety"},{"id":"2604.27426","title":"Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors","created":"2026-04-30","cited_by_count":0,"area":"safety"},{"id":"2605.11612","title":"When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models","created":"2026-05-12","cited_by_count":0,"area":"safety"},{"id":"2605.12890","title":"Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Texts","created":"2026-05-13","cited_by_count":0,"area":"mechinterp"},{"id":"2605.16471","title":"From AI-Generated Content to Agentic Action: Security and Safety Threats in Generative AI","created":"2026-05-15","cited_by_count":0,"area":"agents"},{"id":"2605.26595","title":"Cordyceps: Covert Control Attacks on LLMs via Data Poisoning","created":"2026-05-26","cited_by_count":0,"area":"safety"},{"id":"2605.28597","title":"Position: Retire the \"Positive Backdoor\" Label -- Secret Alignment Requires Strict and Systematic Evaluation","created":"2026-05-27","cited_by_count":0,"area":"alignment"},{"id":"2606.04929","title":"Sequential Data Poisoning in LLM Post-Training","created":"2026-06-03","cited_by_count":0,"area":"alignment"}],"next_cursor":"100"}