{"total":656,"rows":[{"id":"2409.19256","title":"HybridFlow: A Flexible and Efficient RLHF Framework","created":"2024-09-28","cited_by_count":2057,"area":"alignment"},{"id":"2405.14734","title":"SimPO: Simple Preference Optimization with a Reference-Free Reward","created":"2024-05-23","cited_by_count":963,"area":"alignment"},{"id":"2403.13787","title":"RewardBench: Evaluating Reward Models for Language Modeling","created":"2024-03-20","cited_by_count":499,"area":"alignment"},{"id":"2406.18495","title":"WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs","created":"2024-06-26","cited_by_count":496,"area":"safety"},{"id":"2401.10019","title":"R-Judge: Benchmarking Safety Risk Awareness for LLM Agents","created":"2024-01-18","cited_by_count":288,"area":"agents"},{"id":"2404.10719","title":"Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study","created":"2024-04-16","cited_by_count":262,"area":"alignment"},{"id":"2406.15513","title":"PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference","created":"2024-06-20","cited_by_count":261,"area":"alignment"},{"id":"2405.20947","title":"OR-Bench: An Over-Refusal Benchmark for Large Language Models","created":"2024-05-31","cited_by_count":244,"area":"alignment"},{"id":"2406.14598","title":"SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal","created":"2024-06-20","cited_by_count":240,"area":"alignment"},{"id":"2402.05162","title":"Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications","created":"2024-02-07","cited_by_count":231,"area":"alignment"},{"id":"2402.05044","title":"SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Models","created":"2024-02-07","cited_by_count":228,"area":null},{"id":"2406.18510","title":"WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models","created":"2024-06-26","cited_by_count":223,"area":"safety"},{"id":"2601.05242","title":"GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization","created":"2026-01-08","cited_by_count":193,"area":"lr"},{"id":"2404.07921","title":"AmpleGCG: Learning a Universal and Transferable Generative Model of Adversarial Suffixes for Jailbreaking Both Open and Closed LLMs","created":"2024-04-11","cited_by_count":172,"area":"safety"},{"id":"2408.10075","title":"Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning","created":"2024-08-19","cited_by_count":155,"area":"alignment"},{"id":"2403.16950","title":"Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators","created":"2024-03-25","cited_by_count":149,"area":"alignment"},{"id":"2408.17003","title":"Safety Layers in Aligned Large Language Models: The Key to LLM Security","created":"2024-08-30","cited_by_count":134,"area":"alignment"},{"id":"2312.08358","title":"Distributional Preference Learning: Understanding and Accounting for Hidden Context in RLHF","created":"2023-12-13","cited_by_count":133,"area":"alignment"},{"id":"2411.11581","title":"OASIS: Open Agent Social Interaction Simulations with One Million Agents","created":"2024-11-18","cited_by_count":132,"area":null},{"id":"2402.09283","title":"Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey","created":"2024-02-14","cited_by_count":130,"area":null},{"id":"2504.01990","title":"Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems","created":"2025-03-31","cited_by_count":126,"area":"agents"},{"id":"2403.09572","title":"Eyes Closed, Safety On: Protecting Multimodal LLMs via Image-to-Text Transformation","created":"2024-03-14","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":"2407.18418","title":"Know Your Limits: A Survey of Abstention in Large Language Models","created":"2024-07-25","cited_by_count":121,"area":null},{"id":"2401.02906","title":"MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance","created":"2024-01-05","cited_by_count":117,"area":"mllm"},{"id":"2402.01109","title":"Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attack","created":"2024-02-02","cited_by_count":115,"area":"alignment"},{"id":"2506.02177","title":"Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts","created":"2025-06-02","cited_by_count":113,"area":"lr"},{"id":"2411.01111","title":"Rule Based Rewards for Language Model Safety","created":"2024-11-02","cited_by_count":108,"area":"alignment"},{"id":"2401.17256","title":"Weak-to-Strong Jailbreaking on Large Language Models","created":"2024-01-30","cited_by_count":107,"area":"safety"},{"id":"2407.12043","title":"The Art of Saying No: Contextual Noncompliance in Language Models","created":"2024-07-02","cited_by_count":105,"area":null},{"id":"2501.18492","title":"GuardReasoner: Towards Reasoning-based LLM Safeguards","created":"2025-01-30","cited_by_count":100,"area":"lr"},{"id":"2402.02416","title":"Aligner: Efficient Alignment by Learning to Correct","created":"2024-02-04","cited_by_count":98,"area":"alignment"},{"id":"2405.20304","title":"Group Robust Preference Optimization in Reward-free RLHF","created":"2024-05-30","cited_by_count":95,"area":"lr"},{"id":"2407.06551","title":"OffsetBias: Leveraging Debiased Data for Tuning Evaluators","created":"2024-07-09","cited_by_count":94,"area":null},{"id":"2405.18641","title":"Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack","created":"2024-05-28","cited_by_count":92,"area":"alignment"},{"id":"2409.01586","title":"Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation","created":"2024-09-03","cited_by_count":92,"area":null},{"id":"2404.08555","title":"RLHF Deciphered: A Critical Analysis of Reinforcement Learning from Human Feedback for LLMs","created":"2024-04-12","cited_by_count":90,"area":"alignment"},{"id":"2409.18169","title":"Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey","created":"2024-09-26","cited_by_count":90,"area":"alignment"},{"id":"2502.18770","title":"Reward Shaping to Mitigate Reward Hacking in RLHF","created":"2025-02-26","cited_by_count":88,"area":"alignment"},{"id":"2402.02030","title":"Panacea: Pareto Alignment via Preference Adaptation for LLMs","created":"2024-02-03","cited_by_count":80,"area":"alignment"},{"id":"2410.02832","title":"FlipAttack: Jailbreak LLMs via Flipping","created":"2024-10-02","cited_by_count":80,"area":"safety"},{"id":"2502.02384","title":"STAIR: Improving Safety Alignment with Introspective Reasoning","created":"2025-02-04","cited_by_count":77,"area":"lr"},{"id":"2311.08370","title":"SimpleSafetyTests: a Test Suite for Identifying Critical Safety Risks in Large Language Models","created":"2023-11-14","cited_by_count":76,"area":null},{"id":"2501.12368","title":"InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model","created":"2025-01-21","cited_by_count":74,"area":"lr"},{"id":"2408.09600","title":"Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning","created":"2024-08-18","cited_by_count":72,"area":"alignment"},{"id":"2410.13708","title":"On the Role of Attention Heads in Large Language Model Safety","created":"2024-10-17","cited_by_count":72,"area":"safety"},{"id":"2503.03480","title":"SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning","created":"2025-03-05","cited_by_count":72,"area":"vla"},{"id":"2411.16646","title":"Self-Generated Critiques Boost Reward Modeling for Language Models","created":"2024-11-25","cited_by_count":71,"area":"lr"},{"id":"2412.10400","title":"Reinforcement Learning Enhanced LLMs: A Survey","created":"2024-12-05","cited_by_count":71,"area":"lr"},{"id":"2406.18682","title":"The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm","created":"2024-06-26","cited_by_count":70,"area":"safety"},{"id":"2404.05692","title":"Evaluating Mathematical Reasoning Beyond Accuracy","created":"2024-04-08","cited_by_count":69,"area":"lr"},{"id":"2503.24370","title":"Effectively Controlling Reasoning Models through Thinking Intervention","created":"2025-03-31","cited_by_count":65,"area":"lr"},{"id":"2406.05954","title":"Aligning Large Language Models with Representation Editing: A Control Perspective","created":"2024-06-10","cited_by_count":64,"area":"alignment"},{"id":"2410.04070","title":"PAD: Personalized Alignment of LLMs at Decoding-Time","created":"2024-10-05","cited_by_count":60,"area":"alignment"},{"id":"2407.00128","title":"When Search Engine Services meet Large Language Models: Visions and Challenges","created":"2024-06-28","cited_by_count":59,"area":null},{"id":"2504.12501","title":"Reinforcement Learning from Human Feedback","created":"2025-04-16","cited_by_count":58,"area":"alignment"},{"id":"2408.02544","title":"Caution for the Environment: Multimodal LLM Agents are Susceptible to Environmental Distractions","created":"2024-08-05","cited_by_count":56,"area":"mllm"},{"id":"2410.08193","title":"GenARM: Reward Guided Generation with Autoregressive Reward Model for Test-time Alignment","created":"2024-10-10","cited_by_count":54,"area":"alignment"},{"id":"2406.06144","title":"Language Models Resist Alignment: Evidence From Data Compression","created":"2024-06-10","cited_by_count":53,"area":null},{"id":"2406.14144","title":"Towards Understanding Safety Alignment: A Mechanistic Perspective from Safety Neurons","created":"2024-06-20","cited_by_count":53,"area":"alignment"},{"id":"2504.12328","title":"A Comprehensive Survey of Reward Models: Taxonomy, Applications, Challenges, and Future","created":"2025-04-12","cited_by_count":53,"area":"alignment"},{"id":"2507.04632","title":"Can Prompt Difficulty be Online Predicted for Accelerating RL Finetuning of Reasoning Models?","created":"2025-07-07","cited_by_count":52,"area":"lr"},{"id":"2508.09224","title":"From Hard Refusals to Safe-Completions: Toward Output-Centric Safety Training","created":"2025-08-12","cited_by_count":52,"area":null},{"id":"2311.10766","title":"Value FULCRA: Mapping Large Language Models to the Multidimensional Spectrum of Basic Human Values","created":"2023-11-15","cited_by_count":50,"area":null},{"id":"2402.04247","title":"Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy","created":"2024-02-06","cited_by_count":49,"area":null},{"id":"2410.03415","title":"Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation","created":"2024-10-04","cited_by_count":48,"area":null},{"id":"2510.07284","title":"Online Rubrics Elicitation from Pairwise Comparisons","created":"2025-10-08","cited_by_count":46,"area":null},{"id":"2405.14314","title":"Towards Efficient LLM Grounding for Embodied Multi-Agent Collaboration","created":"2024-05-23","cited_by_count":45,"area":"lr"},{"id":"2402.02025","title":"A Survey of Constraint Formulations in Safe Reinforcement Learning","created":"2024-02-03","cited_by_count":44,"area":"deeprl"},{"id":"2405.19358","title":"Robustifying Safety-Aligned Large Language Models through Clean Data Curation","created":"2024-05-24","cited_by_count":44,"area":"alignment"},{"id":"2406.14563","title":"Model Merging and Safety Alignment: One Bad Model Spoils the Bunch","created":"2024-06-20","cited_by_count":43,"area":"alignment"},{"id":"2410.04350","title":"TIS-DPO: Token-level Importance Sampling for Direct Preference Optimization With Estimated Weights","created":"2024-10-06","cited_by_count":42,"area":"alignment"},{"id":"2501.01765","title":"SaLoRA: Safety-Alignment Preserved Low-Rank Adaptation","created":"2025-01-03","cited_by_count":42,"area":"safety"},{"id":"2503.17003","title":"A Survey on Personalized Alignment -- The Missing Piece for Large Language Models in Real-World Applications","created":"2025-03-21","cited_by_count":42,"area":"alignment"},{"id":"2505.11049","title":"GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning","created":"2025-05-16","cited_by_count":42,"area":"lr"},{"id":"2405.09055","title":"A safety realignment framework via subspace-oriented model fusion for large language models","created":"2024-05-15","cited_by_count":41,"area":"safety"},{"id":"2410.12896","title":"A Survey on Data Synthesis and Augmentation for Large Language Models","created":"2024-10-16","cited_by_count":41,"area":"synth"},{"id":"2406.16306","title":"Cascade Reward Sampling for Efficient Decoding-Time Alignment","created":"2024-06-24","cited_by_count":39,"area":"alignment"},{"id":"2502.12668","title":"Evaluation of Best-of-N Sampling Strategies for Language Model Alignment","created":"2025-02-18","cited_by_count":39,"area":"alignment"},{"id":"2506.07468","title":"Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Models","created":"2025-06-09","cited_by_count":39,"area":"alignment"},{"id":"2505.17306","title":"Refusal Direction is Universal Across Safety-Aligned Languages","created":"2025-05-22","cited_by_count":38,"area":null},{"id":"2505.20065","title":"SafeDPO: A Simple Approach to Direct Preference Optimization with Enhanced Safety","created":"2025-05-26","cited_by_count":37,"area":"alignment"},{"id":"2510.12633","title":"Laminar: A Scalable Asynchronous RL Post-Training Framework","created":"2025-10-14","cited_by_count":37,"area":null},{"id":"2310.15469","title":"The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks","created":"2023-10-24","cited_by_count":36,"area":null},{"id":"2403.17830","title":"Assessment of Multimodal Large Language Models in Alignment with Human Values","created":"2024-03-26","cited_by_count":36,"area":"mllm"},{"id":"2408.11779","title":"Personality Alignment of Large Language Models","created":"2024-08-21","cited_by_count":36,"area":"alignment"},{"id":"2406.14477","title":"SafeSora: Towards Safety Alignment of Text2Video Generation via a Human Preference Dataset","created":"2024-06-20","cited_by_count":35,"area":"alignment"},{"id":"2409.00598","title":"Automatic Pseudo-Harmful Prompt Generation for Evaluating False Refusals in Large Language Models","created":"2024-09-01","cited_by_count":35,"area":"alignment"},{"id":"2412.13879","title":"Crabs: Consuming Resource via Auto-generation for LLM-DoS Attack under Black-box Settings","created":"2024-12-18","cited_by_count":35,"area":null},{"id":"2312.04828","title":"HuRef: HUman-REadable Fingerprint for Large Language Models","created":"2023-12-08","cited_by_count":34,"area":null},{"id":"2502.12970","title":"Reasoning-to-Defend: Safety-Aware Reasoning Can Defend Large Language Models from Jailbreaking","created":"2025-02-18","cited_by_count":34,"area":"lr"},{"id":"2505.22411","title":"Mitigating Overthinking in Large Reasoning Models via Manifold Steering","created":"2025-05-28","cited_by_count":33,"area":"lr"},{"id":"2604.13602","title":"Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges","created":"2026-04-15","cited_by_count":33,"area":"alignment"},{"id":"2311.09641","title":"RLHFPoison: Reward Poisoning Attack for Reinforcement Learning with Human Feedback in Large Language Models","created":"2023-11-16","cited_by_count":32,"area":"alignment"},{"id":"2402.11889","title":"ROSE Doesn't Do That: Boosting the Safety of Instruction-Tuned Large Language Models with Reverse Prompt Contrastive Decoding","created":"2024-02-19","cited_by_count":32,"area":null},{"id":"2402.16006","title":"ASETF: A Novel Method for Jailbreak Attack on LLMs through Translate Suffix Embeddings","created":"2024-02-25","cited_by_count":32,"area":"safety"},{"id":"2405.17888","title":"Getting More Juice Out of the SFT Data: Reward Learning from Human Demonstration Improves SFT for LLM Alignment","created":"2024-05-28","cited_by_count":32,"area":"alignment"},{"id":"2409.14586","title":"Backtracking Improves Generation Safety","created":"2024-09-22","cited_by_count":32,"area":"alignment"},{"id":"2505.17147","title":"MTSA: Multi-turn Safety Alignment for LLMs through Multi-round Red-teaming","created":"2025-05-22","cited_by_count":32,"area":"alignment"},{"id":"2404.11049","title":"Stepwise Alignment for Constrained Language Model Policy Optimization","created":"2024-04-17","cited_by_count":30,"area":"alignment"}],"next_cursor":"100"}