{"total":43,"rows":[{"id":"2412.14093","title":"Alignment faking in large language models","created":"2024-12-18","cited_by_count":360,"area":"alignment"},{"id":"2511.18397","title":"Natural Emergent Misalignment from Reward Hacking in Production RL","created":"2025-11-23","cited_by_count":136,"area":"alignment"},{"id":"2409.12822","title":"Language Models Learn to Mislead Humans via RLHF","created":"2024-09-19","cited_by_count":124,"area":"alignment"},{"id":"2409.13156","title":"RRM: Robust Reward Model Training Mitigates Reward Hacking","created":"2024-09-20","cited_by_count":76,"area":"alignment"},{"id":"2508.17511","title":"School of Reward Hacks: Hacking harmless tasks generalizes to misaligned behavior in LLMs","created":"2025-08-24","cited_by_count":66,"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":"2409.11704","title":"From Lists to Emojis: How Format Bias Affects Model Alignment","created":"2024-09-18","cited_by_count":49,"area":"alignment"},{"id":"2411.02306","title":"On Targeted Manipulation and Deception when Optimizing LLMs for User Feedback","created":"2024-11-04","cited_by_count":47,"area":"alignment"},{"id":"2501.09620","title":"Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment","created":"2025-01-16","cited_by_count":42,"area":"alignment"},{"id":"2506.19248","title":"Inference-Time Reward Hacking in Large Language Models","created":"2025-06-24","cited_by_count":34,"area":"alignment"},{"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":"2406.11431","title":"Super(ficial)-alignment: Strong Models May Deceive Weak Models in Weak-to-Strong Generalization","created":"2024-06-17","cited_by_count":23,"area":"alignment"},{"id":"2501.13011","title":"MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking","created":"2025-01-22","cited_by_count":19,"area":"alignment"},{"id":"2506.16507","title":"Robust Reward Modeling via Causal Rubrics","created":"2025-06-19","cited_by_count":19,"area":"alignment"},{"id":"2501.08617","title":"RLHS: Mitigating Misalignment in RLHF with Hindsight Simulation","created":"2025-01-15","cited_by_count":15,"area":"alignment"},{"id":"2601.22664","title":"Real-Time Aligned Reward Model beyond Semantics","created":"2026-01-30","cited_by_count":13,"area":"alignment"},{"id":"2406.14393","title":"Jailbreaking as a Reward Misspecification Problem","created":"2024-06-20","cited_by_count":11,"area":"alignment"},{"id":"2602.01750","title":"Adversarial Reward Auditing for Active Detection and Mitigation of Reward Hacking","created":"2026-02-02","cited_by_count":11,"area":"alignment"},{"id":"2605.20744","title":"Hack-Verifiable Environments: Towards Evaluating Reward Hacking at Scale","created":"2026-05-20","cited_by_count":10,"area":"alignment"},{"id":"2410.06491","title":"Honesty to Subterfuge: In-Context Reinforcement Learning Can Make Honest Models Reward Hack","created":"2024-10-09","cited_by_count":8,"area":"alignment"},{"id":"2512.23461","title":"Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance","created":"2025-12-29","cited_by_count":7,"area":"alignment"},{"id":"2604.01476","title":"From Rebound to Remedy: Understanding and Mitigating Reward Hacking via Representation Engineering","created":"2026-04-01","cited_by_count":7,"area":"alignment"},{"id":"2406.15753","title":"The Perils of Optimizing Learned Reward Functions: Low Training Error Does Not Guarantee Low Regret","created":"2024-06-22","cited_by_count":6,"area":"alignment"},{"id":"2502.14619","title":"Reward Models Identify Consistency, Not Causality","created":"2025-02-20","cited_by_count":6,"area":"alignment"},{"id":"2507.18742","title":"Specification Self-Correction: Mitigating In-Context Reward Hacking Through Test-Time Refinement","created":"2025-07-24","cited_by_count":6,"area":"alignment"},{"id":"2510.19050","title":"Rectifying Shortcut Behaviors in Preference-based Reward Learning","created":"2025-10-21","cited_by_count":6,"area":"alignment"},{"id":"2411.10156","title":"Mitigating Sycophancy in Decoder-Only Transformer Architectures: Synthetic Data Intervention","created":"2024-11-15","cited_by_count":5,"area":"alignment"},{"id":"2604.16242","title":"Detecting and Suppressing Reward Hacking with Gradient Fingerprints","created":"2026-04-17","cited_by_count":5,"area":"alignment"},{"id":"2604.02686","title":"Beyond Semantic Manipulation: Token-Space Attacks on Reward Models","created":"2026-04-03","cited_by_count":3,"area":"alignment"},{"id":"2607.18966","title":"Measuring Reward-Seeking via Contrastive Belief Updates","created":"2026-07-21","cited_by_count":3,"area":"alignment"},{"id":"2509.21798","title":"Evaluating and Improving Cultural Awareness of Reward Models for LLM Alignment","created":"2025-09-26","cited_by_count":2,"area":"alignment"},{"id":"2506.00152","title":"Aligning Language Models with Observational Data: Opportunities and Risks from a Causal Perspective","created":"2025-05-30","cited_by_count":1,"area":"alignment"},{"id":"2604.26511","title":"Tatemae: Detecting Alignment Faking via Tool Selection in LLMs","created":"2026-04-29","cited_by_count":1,"area":"alignment"},{"id":"2609.36308","title":"CheatBench: Measuring Reward Gaming in AI Agents","created":"2026-09-28","cited_by_count":1,"area":"alignment"},{"id":"2610.10947","title":"RH-Detect: A Unified Benchmark for Reward Hacking Detection","created":"2026-10-07","cited_by_count":1,"area":"alignment"},{"id":"2508.06592","title":"Towards Integrated Alignment","created":"2025-08-08","cited_by_count":0,"area":"alignment"},{"id":"2603.29993","title":"Extending MONA in Camera Dropbox: Reproduction, Learned Approval, and Design Implications for Reward-Hacking Mitigation","created":"2026-03-31","cited_by_count":0,"area":"alignment"},{"id":"2604.02585","title":"Mitigating LLM biases toward spurious social contexts using direct preference optimization","created":"2026-04-02","cited_by_count":0,"area":"alignment"},{"id":"2605.27355","title":"Alignment Tampering: How Reinforcement Learning from Human Feedback Is Exploited to Optimize Misaligned Biases","created":"2026-05-26","cited_by_count":0,"area":"alignment"},{"id":"2606.07532","title":"Durable Evaluation Framework: Adversarial Arbitration for Sycophancy Reduction in Large Language Models","created":"2026-04-21","cited_by_count":0,"area":"alignment"},{"id":"2606.28955","title":"Modification-Considering Value Learning for Reward Hacking Mitigation in RL","created":"2026-06-27","cited_by_count":0,"area":"alignment"},{"id":"2609.35677","title":"Verifier Errors in RLVR: Reward Hacking, Limits of Feedback, and Selective Control","created":"2026-09-28","cited_by_count":0,"area":"alignment"},{"id":"2610.03458","title":"A Near-Zero Monitor Readout Is Not Evidence of Behavioral Control","created":"2026-10-02","cited_by_count":0,"area":"alignment"}],"next_cursor":null}