{"total":78,"rows":[{"id":"2505.11475","title":"HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languages","created":"2025-05-16","cited_by_count":75,"area":"alignment"},{"id":"2501.12895","title":"Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback","created":"2025-01-22","cited_by_count":41,"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":"2410.08968","title":"Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements","created":"2024-10-11","cited_by_count":36,"area":"alignment"},{"id":"2409.15268","title":"Style Outweighs Substance: Failure Modes of LLM Judges in Alignment Benchmarking","created":"2024-09-23","cited_by_count":35,"area":"alignment"},{"id":"2410.17578","title":"MM-Eval: A Multilingual Meta-Evaluation Benchmark for LLM-as-a-Judge and Reward Models","created":"2024-10-23","cited_by_count":35,"area":"alignment"},{"id":"2506.09996","title":"From Judgment to Interference: Early Stopping LLM Harmful Outputs via Streaming Content Monitoring","created":"2025-06-11","cited_by_count":22,"area":"alignment"},{"id":"2506.10597","title":"SoK: Evaluating Jailbreak Guardrails for Large Language Models","created":"2025-06-12","cited_by_count":21,"area":"alignment"},{"id":"2602.01539","title":"MAGIC: A Co-Evolving Attacker-Defender Adversarial Game for Robust LLM Safety","created":"2026-02-02","cited_by_count":21,"area":"alignment"},{"id":"2506.16507","title":"Robust Reward Modeling via Causal Rubrics","created":"2025-06-19","cited_by_count":19,"area":"alignment"},{"id":"2506.17368","title":"SAFEx: Analyzing Vulnerabilities of MoE-Based LLMs via Stable Safety-critical Expert Identification","created":"2025-06-20","cited_by_count":18,"area":"alignment"},{"id":"2410.18027","title":"Cross-lingual Transfer of Reward Models in Multilingual Alignment","created":"2024-10-23","cited_by_count":17,"area":"alignment"},{"id":"2507.14987","title":"AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning","created":"2025-07-20","cited_by_count":15,"area":"alignment"},{"id":"2508.06249","title":"In-Training Defenses against Emergent Misalignment in Language Models","created":"2025-08-08","cited_by_count":15,"area":"alignment"},{"id":"2505.21605","title":"SoSBench: Benchmarking Safety Alignment on Six Scientific Domains","created":"2025-05-27","cited_by_count":12,"area":"alignment"},{"id":"2505.12763","title":"Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization","created":"2025-05-19","cited_by_count":11,"area":"alignment"},{"id":"2506.21463","title":"Aligning Spoken Dialogue Models from User Interactions","created":"2025-06-26","cited_by_count":11,"area":"alignment"},{"id":"2601.07349","title":"Reward Modeling from Natural Language Human Feedback","created":"2026-01-12","cited_by_count":11,"area":"alignment"},{"id":"2512.20806","title":"Safety Alignment of LMs via Non-cooperative Games","created":"2025-12-23","cited_by_count":8,"area":"alignment"},{"id":"2603.18736","title":"CausalRM: Causal-Theoretic Reward Modeling for RLHF from Observational User Feedbacks","created":"2026-03-19","cited_by_count":7,"area":"alignment"},{"id":"2506.00166","title":"Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignment","created":"2025-05-30","cited_by_count":6,"area":"alignment"},{"id":"2507.00665","title":"SAFER: Probing Safety in Reward Models with Sparse Autoencoder","created":"2025-07-01","cited_by_count":6,"area":"alignment"},{"id":"2509.06982","title":"CARE: Decoding Time Safety Alignment via Rollback and Introspection Intervention","created":"2025-09-01","cited_by_count":6,"area":"alignment"},{"id":"2501.13264","title":"OpenGenAlign: A Preference Dataset and Benchmark for Trustworthy Reward Modeling in Open-Ended, Long-Context Generation","created":"2025-01-22","cited_by_count":5,"area":"alignment"},{"id":"2508.20766","title":"Turning the Spell Around: Lightweight Alignment Amplification via Rank-One Safety Injection","created":"2025-08-28","cited_by_count":5,"area":"alignment"},{"id":"2510.08646","title":"Mitigating Over-Refusal in Aligned Large Language Models via Inference-Time Activation Energy","created":"2025-10-09","cited_by_count":5,"area":"alignment"},{"id":"2601.17642","title":"Health-ORSC-Bench: A Benchmark for Measuring Over-Refusal and Safety Completion in Health Context","created":"2026-01-25","cited_by_count":4,"area":"alignment"},{"id":"2604.18901","title":"Harmful Intent as a Geometrically Recoverable Feature of LLM Residual Streams","created":"2026-04-20","cited_by_count":4,"area":"alignment"},{"id":"2510.18081","title":"Any-Depth Alignment: Unlocking Innate Safety Alignment of LLMs to Any-Depth","created":"2025-10-20","cited_by_count":3,"area":"alignment"},{"id":"2511.06222","title":"SPA: Achieving Consensus in LLM Alignment via Self-Priority Optimization","created":"2025-11-09","cited_by_count":3,"area":"alignment"},{"id":"2603.23184","title":"ImplicitRM: Unbiased Reward Modeling from Implicit Preference Data for LLM alignment","created":"2026-03-24","cited_by_count":3,"area":"alignment"},{"id":"2604.27093","title":"Useless but Safe? Benchmarking Utility Recovery with User Intent Clarification in Multi-Turn Conversations","created":"2026-04-29","cited_by_count":3,"area":"alignment"},{"id":"2605.31328","title":"Reinforcement Learning Can Amplify Emergent Misalignment from Harmless Rewards","created":"2026-05-29","cited_by_count":3,"area":"alignment"},{"id":"2512.01037","title":"How Semantically Stable Are LLM Refusals? Measuring Confusion in Local Safety Boundaries","created":"2025-11-30","cited_by_count":2,"area":"alignment"},{"id":"2603.10243","title":"GR-SAP: Generative Replay for Safety Alignment Preservation during Fine-Tuning","created":"2026-03-10","cited_by_count":2,"area":"alignment"},{"id":"2603.12795","title":"SteerRM: Debiasing Reward Models via Sparse Autoencoders","created":"2026-03-13","cited_by_count":2,"area":"alignment"},{"id":"2604.17132","title":"Please refuse to answer me! Mitigating Over-Refusal in Large Language Models via Adaptive Contrastive Decoding","created":"2026-04-18","cited_by_count":2,"area":"alignment"},{"id":"2604.26167","title":"Test-Time Safety Alignment","created":"2026-04-28","cited_by_count":2,"area":"alignment"},{"id":"2511.00509","title":"Reimagining Safety Alignment with An Image","created":"2025-11-01","cited_by_count":1,"area":"alignment"},{"id":"2602.03073","title":"TMS: Trajectory-Mixed Supervision for Reward-Free, On-Policy SFT","created":"2026-02-03","cited_by_count":1,"area":"alignment"},{"id":"2603.13359","title":"From Refusal Tokens to Refusal Control: Discovering and Steering Category-Specific Refusal Directions","created":"2026-03-09","cited_by_count":1,"area":"alignment"},{"id":"2604.09189","title":"Do LLMs Follow Their Own Rules? A Reflexive Audit of Self-Stated Safety Policies","created":"2026-04-10","cited_by_count":1,"area":"alignment"},{"id":"2605.01899","title":"Disentangling Intent from Role: Adversarial Self-Play for Persona-Invariant Safety Alignment","created":"2026-05-03","cited_by_count":1,"area":"alignment"},{"id":"2606.19168","title":"Beyond Safe Data: Pretraining-Stage Alignment with Regular Safety Reflection","created":"2026-06-17","cited_by_count":1,"area":"alignment"},{"id":"2606.28153","title":"Robust Harmful Features Under Jailbreak Attacks: Mechanistic Evidence from Attention Head Specialization in Large Language Models","created":"2026-06-26","cited_by_count":1,"area":"alignment"},{"id":"2608.12645","title":"Jagged Judges: Epistemic Stability Under Perturbation, Pressure, and Persistence","created":"2026-08-12","cited_by_count":1,"area":"alignment"},{"id":"2608.16577","title":"BabelSteering: Multilingual Safety Alignment via English Steering Vectors","created":"2026-08-17","cited_by_count":1,"area":"alignment"},{"id":"2508.09937","title":"A Comprehensive Evaluation framework of Alignment Techniques for LLMs","created":"2025-08-13","cited_by_count":0,"area":"alignment"},{"id":"2511.13548","title":"ForgeDAN: An Evolutionary Framework for Jailbreaking Aligned Large Language Models","created":"2025-11-17","cited_by_count":0,"area":"alignment"},{"id":"2601.04262","title":"Safety-Utility Conflicts Are Not Global: Surgical Alignment via Head-Level Diagnosis","created":"2026-01-07","cited_by_count":0,"area":"alignment"},{"id":"2601.04389","title":"Safety Is Not Universal: The Selective Safety Trap in LLM Alignment","created":"2026-01-07","cited_by_count":0,"area":"alignment"},{"id":"2602.04294","title":"How Few-shot Demonstrations Affect Prompt-based Defenses Against LLM Jailbreak Attacks","created":"2026-02-04","cited_by_count":0,"area":"alignment"},{"id":"2602.08819","title":"Bayesian Preference Learning for Test-Time Steerable Reward Models","created":"2026-02-09","cited_by_count":0,"area":"alignment"},{"id":"2603.15647","title":"Steering Frozen LLMs: Adaptive Social Alignment via Online Prompt Routing","created":"2026-03-04","cited_by_count":0,"area":"alignment"},{"id":"2604.17691","title":"SafeAnchor: Preventing Cumulative Safety Erosion in Continual Domain Adaptation of Large Language Models","created":"2026-04-20","cited_by_count":0,"area":"alignment"},{"id":"2605.09225","title":"The Art of the Jailbreak: Formulating Jailbreak Attacks for LLM Security Beyond Binary Scoring","created":"2026-05-09","cited_by_count":0,"area":"alignment"},{"id":"2605.10998","title":"Few-Shot Truly Benign DPO Attack for Jailbreaking LLMs","created":"2026-05-09","cited_by_count":0,"area":"alignment"},{"id":"2605.11712","title":"Toward Stable Value Alignment: Introducing Independent Modules for Consistent Value Guidance","created":"2026-05-12","cited_by_count":0,"area":"alignment"},{"id":"2605.30487","title":"Configurable Reward Model for Balanced Safety Alignment","created":"2026-05-28","cited_by_count":0,"area":"alignment"},{"id":"2606.15396","title":"CHILLGuard: Towards Fine-Grained Chinese LLM Safety Guardrail with Scalable Data Construction and Model-aware Preference Alignment","created":"2026-06-13","cited_by_count":0,"area":"alignment"},{"id":"2606.20508","title":"What Do Safety-Aligned LLMs Learn From Mixed Compliance Demonstrations?","created":"2026-06-18","cited_by_count":0,"area":"alignment"},{"id":"2606.29581","title":"The Joint Effect of Quantization and Sampling Temperature on LLM Safety Alignment: A Factorial Analysis","created":"2026-06-28","cited_by_count":0,"area":"alignment"},{"id":"2607.00572","title":"HARC: Coupling Harmfulness and Refusal Directions for Robust Safety Alignment","created":"2026-07-01","cited_by_count":0,"area":"alignment"},{"id":"2607.10112","title":"Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety","created":"2026-07-11","cited_by_count":0,"area":"alignment"},{"id":"2607.11475","title":"HyperSafe: Inference-Time Safety Recovery for Fine-Tuned Language Models","created":"2026-07-13","cited_by_count":0,"area":"alignment"},{"id":"2608.05409","title":"Mood Matters: How Syntactic Sensitivity Undermines Safety Alignment","created":"2026-08-05","cited_by_count":0,"area":"alignment"},{"id":"2608.11705","title":"Making Your LLMs More Objective: Stabilizing LLM Safety Behavior Across Traits with Trait-Invariant Safety Tuning","created":"2026-08-12","cited_by_count":0,"area":"alignment"},{"id":"2608.20554","title":"aiXamine: Unified Black-Box Evaluation of Cross-Dimensional Trade-offs in LLM Safety, Security, and Privacy","created":"2026-08-20","cited_by_count":0,"area":"alignment"},{"id":"2609.00699","title":"Patterning in Practice: Debiasing Reward Models with Susceptibilities","created":"2026-09-01","cited_by_count":0,"area":"alignment"},{"id":"2609.02215","title":"ASCII Attack: Recontextualising Harmful Requests as Artistic Critique in Large Language Models","created":"2026-09-02","cited_by_count":0,"area":"alignment"},{"id":"2609.02273","title":"CoMerge: Conflict-Driven Preference Optimization for Multi-Task Model Merging","created":"2026-09-02","cited_by_count":0,"area":"alignment"},{"id":"2609.03693","title":"AlcaTRAz - Anchored Tree-Rule Defense Against Jailbreaks","created":"2026-09-03","cited_by_count":0,"area":"alignment"},{"id":"2609.04482","title":"Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal","created":"2026-09-03","cited_by_count":0,"area":"alignment"},{"id":"2609.04714","title":"Refuse without Refusal: A Structural Analysis of Safety-Tuning Responses for Reducing False Refusals in Language Models","created":"2026-09-04","cited_by_count":0,"area":"alignment"},{"id":"2609.19325","title":"AUDITPLAN: Commit, Then Answer for Auditable Safety Alignment","created":"2026-09-16","cited_by_count":0,"area":"alignment"},{"id":"2609.30802","title":"Understanding the Role of Prompt Template in Knowledge Distillation for Safety Alignment","created":"2026-09-25","cited_by_count":0,"area":"alignment"},{"id":"2609.34519","title":"EOPSA: Efficient On-Policy Self-Distilled Safety Alignment","created":"2026-09-28","cited_by_count":0,"area":"alignment"},{"id":"2609.34771","title":"When Do Model Internals Help? Exploring the Role of Representation Engineering in LLM Safety","created":"2026-09-28","cited_by_count":0,"area":"alignment"}],"next_cursor":null}