{"total":57,"rows":[{"id":"2401.06373","title":"How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs","created":"2024-01-12","cited_by_count":573,"area":"lr"},{"id":"2506.04178","title":"OpenThoughts: Data Recipes for Reasoning Models","created":"2025-06-04","cited_by_count":304,"area":"lr"},{"id":"2402.08787","title":"Rethinking Machine Unlearning for Large Language Models","created":"2024-02-13","cited_by_count":257,"area":"lr"},{"id":"2406.09187","title":"GuardAgent: Safeguard LLM Agents by a Guard Agent via Knowledge-Enabled Reasoning","created":"2024-06-13","cited_by_count":124,"area":"lr"},{"id":"2503.00555","title":"Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable","created":"2025-03-01","cited_by_count":122,"area":"lr"},{"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":"2502.12893","title":"H-CoT: Hijacking the Chain-of-Thought Safety Reasoning Mechanism to Jailbreak Large Reasoning Models, Including OpenAI o1/o3, DeepSeek-R1, and Gemini 2.0 Flash Thinking","created":"2025-02-18","cited_by_count":99,"area":"lr"},{"id":"2502.12659","title":"The Hidden Risks of Large Reasoning Models: A Safety Assessment of R1","created":"2025-02-18","cited_by_count":91,"area":"lr"},{"id":"2504.01903","title":"STAR-1: Safer Alignment of Reasoning LLMs with 1K Data","created":"2025-04-02","cited_by_count":68,"area":"lr"},{"id":"2502.19301","title":"Rethinking LLM Unlearning Objectives: A Gradient Perspective and Go Beyond","created":"2025-02-26","cited_by_count":61,"area":"lr"},{"id":"2505.08054","title":"FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning","created":"2025-05-12","cited_by_count":58,"area":"lr"},{"id":"2405.18634","title":"A Theoretical Understanding of Self-Correction through In-context Alignment","created":"2024-05-28","cited_by_count":44,"area":"lr"},{"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":"2506.13752","title":"Steering LLM Thinking with Budget Guidance","created":"2025-06-16","cited_by_count":33,"area":"lr"},{"id":"2405.20774","title":"Can We Trust Embodied Agents? Exploring Backdoor Attacks against Embodied LLM-based Decision-Making Systems","created":"2024-05-27","cited_by_count":27,"area":"lr"},{"id":"2502.09673","title":"Are Smarter LLMs Safer? Exploring Safety-Reasoning Trade-offs in Prompting and Fine-Tuning","created":"2025-02-13","cited_by_count":20,"area":"lr"},{"id":"2505.16186","title":"SafeKey: Amplifying Aha-Moment Insights for Safety Reasoning","created":"2025-05-22","cited_by_count":20,"area":"lr"},{"id":"2502.15799","title":"Investigating the Impact of Quantization Methods on the Safety and Reliability of Large Language Models","created":"2025-02-18","cited_by_count":18,"area":"lr"},{"id":"2508.12531","title":"Rethinking Safety in LLM Fine-tuning: An Optimization Perspective","created":"2025-08-17","cited_by_count":14,"area":"lr"},{"id":"2502.01225","title":"The dark deep side of DeepSeek: Fine-tuning attacks against the safety alignment of CoT-enabled models","created":"2025-02-03","cited_by_count":12,"area":"lr"},{"id":"2503.17882","title":"Think Before Refusal : Triggering Safety Reflection in LLMs to Mitigate False Refusal Behavior","created":"2025-03-22","cited_by_count":12,"area":"lr"},{"id":"2508.16347","title":"Confusion is the Final Barrier: Rethinking Jailbreak Evaluation and Investigating the Real Misuse Threat of LLMs","created":"2025-08-22","cited_by_count":11,"area":"lr"},{"id":"2512.12069","title":"Rethinking Jailbreak Detection of Large Vision Language Models with Representational Contrastive Scoring","created":"2025-12-12","cited_by_count":11,"area":"lr"},{"id":"2605.15239","title":"Reducing the Safety Tax in LLM Safety Alignment with On-Policy Self-Distillation","created":"2026-05-14","cited_by_count":11,"area":"lr"},{"id":"2504.16489","title":"Amplified Vulnerabilities: Structured Jailbreak Attacks on LLM-based Multi-Agent Debate","created":"2025-04-23","cited_by_count":10,"area":"lr"},{"id":"2510.09714","title":"All Code, No Thought: Current Language Models Struggle to Reason in Ciphered Language","created":"2025-10-10","cited_by_count":8,"area":"lr"},{"id":"2603.01761","title":"Position: Modular Memory is the Key to Continual Learning Agents","created":"2026-03-02","cited_by_count":8,"area":"lr"},{"id":"2506.00668","title":"SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialogues","created":"2025-05-31","cited_by_count":7,"area":"lr"},{"id":"2507.17075","title":"LoRA is All You Need for Safety Alignment of Reasoning LLMs","created":"2025-07-22","cited_by_count":6,"area":"lr"},{"id":"2504.02725","title":"SAFER: Advancing Safety Alignment via Efficient Ex-Ante Reasoning","created":"2025-04-03","cited_by_count":4,"area":"lr"},{"id":"2505.17089","title":"Chain-of-Thought Driven Adversarial Scenario Extrapolation for Robust Language Models","created":"2025-05-20","cited_by_count":4,"area":"lr"},{"id":"2508.04204","title":"ReasoningGuard: Safeguarding Large Reasoning Models with Inference-time Safety Aha Moments","created":"2025-08-06","cited_by_count":4,"area":"lr"},{"id":"2605.21127","title":"Reasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-Tuning","created":"2026-05-20","cited_by_count":4,"area":"lr"},{"id":"2505.21784","title":"Towards Safety Reasoning in LLMs: AI-agentic Deliberation for Policy-embedded CoT Data Creation","created":"2025-05-27","cited_by_count":3,"area":"lr"},{"id":"2601.03537","title":"STAR-S: Improving Safety Alignment through Self-Taught Reasoning on Safety Rules","created":"2026-01-07","cited_by_count":3,"area":"lr"},{"id":"2603.15661","title":"DynaTrust: Defending Multi-Agent Systems Against Sleeper Agents via Dynamic Trust Graphs","created":"2026-03-09","cited_by_count":3,"area":"lr"},{"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":"2602.21346","title":"Alignment-Weighted DPO: A principled reasoning approach to improve safety alignment","created":"2026-02-24","cited_by_count":2,"area":"lr"},{"id":"2604.00012","title":"Finding and Reactivating Post-Trained LLMs' Hidden Safety Mechanisms","created":"2026-03-10","cited_by_count":2,"area":"lr"},{"id":"2501.01872","title":"Turning Logic Against Itself : Probing Model Defenses Through Contrastive Questions","created":"2025-01-03","cited_by_count":1,"area":"lr"},{"id":"2508.03054","title":"Beyond Surface-Level Detection: Towards Cognitive-Driven Defense Against Jailbreak Attacks via Meta-Operations Reasoning","created":"2025-08-05","cited_by_count":1,"area":"lr"},{"id":"2508.10390","title":"Jailbreaking Commercial Black-Box LLMs with Explicitly Harmful Prompts","created":"2025-08-14","cited_by_count":1,"area":"lr"},{"id":"2511.14106","title":"Stealth Fine-Tuning: Efficiently Breaking Alignment in RVLMs Using Self-Generated CoT","created":"2025-11-18","cited_by_count":1,"area":"lr"},{"id":"2604.09750","title":"Conflicts Make Large Reasoning Models Vulnerable to Attacks","created":"2026-04-10","cited_by_count":1,"area":"lr"},{"id":"2608.21727","title":"Reinforcement Learning on Benign Facts Amplifies Leakage of Memorized Private Data","created":"2026-08-22","cited_by_count":1,"area":"lr"},{"id":"2509.05367","title":"Between a Rock and a Hard Place: The Tension Between Ethical Reasoning and Safety Alignment in LLMs","created":"2025-09-04","cited_by_count":0,"area":"lr"},{"id":"2511.12155","title":"Rethinking Deep Alignment Through The Lens Of Incomplete Learning","created":"2025-11-15","cited_by_count":0,"area":"lr"},{"id":"2603.02675","title":"From Shallow to Deep: Pinning Semantic Intent via Causal GRPO","created":"2026-03-03","cited_by_count":0,"area":"lr"},{"id":"2604.04120","title":"Shorter, but Still Trustworthy? An Empirical Study of Chain-of-Thought Compression","created":"2026-04-05","cited_by_count":0,"area":"lr"},{"id":"2604.12088","title":"Think Before You Code: Dual Reasoning for the NLSafety-Utility Trade-Off in LLM Code Generation","created":"2026-04-13","cited_by_count":0,"area":"lr"},{"id":"2604.17429","title":"Jupiter-N Technical Report","created":"2026-04-19","cited_by_count":0,"area":"lr"},{"id":"2605.15384","title":"Is One Score Enough? Rethinking the Evaluation of Sequentially Evolving LLM Memory","created":"2026-05-14","cited_by_count":0,"area":"lr"},{"id":"2606.11046","title":"Does Reasoning Preserve Alignment? On the Trustworthiness of Large Reasoning Models","created":"2026-06-09","cited_by_count":0,"area":"lr"},{"id":"2606.16808","title":"Adaptive and Explicit safe: Triggering Latent Safety Awareness in Large Reasoning Models","created":"2026-06-15","cited_by_count":0,"area":"lr"},{"id":"2607.28146","title":"Can Agents Deceive? Evaluating Reasoning and Deception in ParliamentBench using a Social Deduction Game","created":"2026-07-30","cited_by_count":0,"area":"lr"},{"id":"2608.23497","title":"Mitigating Reasoning-Induced Misalignment via Safety-Direction Penalty","created":"2026-08-24","cited_by_count":0,"area":"lr"},{"id":"2610.16312","title":"Narrative Paradoxes in LLM Inference: Shaping Reasoning Trajectories toward Misalignment","created":"2026-10-12","cited_by_count":0,"area":"lr"}],"next_cursor":null}