{"total":66,"rows":[{"id":"2406.05946","title":"Safety Alignment Should Be Made More Than Just a Few Tokens Deep","created":"2024-06-10","cited_by_count":472,"area":"alignment"},{"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":"2312.07000","title":"Alignment for Honesty","created":"2023-12-12","cited_by_count":74,"area":"alignment"},{"id":"2402.15018","title":"Unintended Impacts of LLM Alignment on Global Representation","created":"2024-02-22","cited_by_count":66,"area":"alignment"},{"id":"2410.03642","title":"Aligning LLMs with Individual Preferences via Interaction","created":"2024-10-04","cited_by_count":63,"area":"alignment"},{"id":"2406.16768","title":"WARP: On the Benefits of Weight Averaged Rewarded Policies","created":"2024-06-24","cited_by_count":45,"area":"alignment"},{"id":"2404.16792","title":"Model Extrapolation Expedites Alignment","created":"2024-04-25","cited_by_count":37,"area":"alignment"},{"id":"2503.11701","title":"A Survey of Direct Preference Optimization","created":"2025-03-12","cited_by_count":36,"area":"alignment"},{"id":"2406.05587","title":"Creativity Has Left the Chat: The Price of Debiasing Language Models","created":"2024-06-08","cited_by_count":29,"area":"alignment"},{"id":"2509.03403","title":"Beyond Correctness: Harmonizing Process and Outcome Rewards through RL Training","created":"2025-09-03","cited_by_count":29,"area":"alignment"},{"id":"2405.17931","title":"Online Merging Optimizers for Boosting Rewards and Mitigating Tax in Alignment","created":"2024-05-28","cited_by_count":28,"area":"alignment"},{"id":"2407.08770","title":"Model Surgery: Modulating LLM's Behavior Via Simple Parameter Editing","created":"2024-07-11","cited_by_count":23,"area":"alignment"},{"id":"2503.18991","title":"Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment","created":"2025-03-23","cited_by_count":20,"area":"alignment"},{"id":"2406.10991","title":"Adaptive Query Rewriting: Aligning Rewriters through Marginal Probability of Conversational Answers","created":"2024-06-16","cited_by_count":15,"area":"alignment"},{"id":"2507.21183","title":"MaPPO: Maximum a Posteriori Preference Optimization with Prior Knowledge","created":"2025-07-27","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":"2410.10862","title":"Superficial Safety Alignment Hypothesis","created":"2024-10-07","cited_by_count":14,"area":"alignment"},{"id":"2507.07375","title":"Bradley-Terry and Multi-Objective Reward Modeling Are Complementary","created":"2025-07-10","cited_by_count":11,"area":"alignment"},{"id":"2512.11391","title":"Mitigating the Safety Alignment Tax with Null-Space Constrained Policy Optimization","created":"2025-12-12","cited_by_count":11,"area":"alignment"},{"id":"2510.17426","title":"Navigating the Alignment-Calibration Trade-off: A Pareto-Superior Frontier via Model Merging","created":"2025-10-20","cited_by_count":10,"area":"alignment"},{"id":"2501.06481","title":"Focus-N-Fix: Region-Aware Fine-Tuning for Text-to-Image Generation","created":"2025-01-11","cited_by_count":9,"area":"alignment"},{"id":"2502.00666","title":"Avoiding $\\mathbf{exp(R_{max})}$ scaling in RLHF through Preference-based Exploration","created":"2025-02-02","cited_by_count":8,"area":"alignment"},{"id":"2507.11500","title":"ARMOR: Aligning Secure and Safe Large Language Models via Meticulous Reasoning","created":"2025-07-14","cited_by_count":8,"area":"alignment"},{"id":"2407.03856","title":"Q-Adapter: Customizing Pre-trained LLMs to New Preferences with Forgetting Mitigation","created":"2024-07-04","cited_by_count":7,"area":"alignment"},{"id":"2509.24610","title":"OrthAlign: Orthogonal Subspace Decomposition for Non-Interfering Multi-Objective Alignment","created":"2025-09-29","cited_by_count":7,"area":"alignment"},{"id":"2410.04834","title":"As Simple as Fine-tuning: LLM Alignment via Bidirectional Negative Feedback Loss","created":"2024-10-07","cited_by_count":6,"area":"alignment"},{"id":"2602.07892","title":"Safety Alignment as Continual Learning: Mitigating the Alignment Tax via Orthogonal Gradient Projection","created":"2026-02-08","cited_by_count":6,"area":"alignment"},{"id":"2509.05381","title":"Murphys Laws of AI Alignment: Why the Gap Always Wins","created":"2025-09-04","cited_by_count":5,"area":"alignment"},{"id":"2508.09473","title":"NeuronTune: Fine-Grained Neuron Modulation for Balanced Safety-Utility Alignment in LLMs","created":"2025-08-13","cited_by_count":4,"area":"alignment"},{"id":"2509.23765","title":"Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality","created":"2025-09-28","cited_by_count":4,"area":"alignment"},{"id":"2510.06652","title":"Aligning Large Language Models via Fully Self-Synthetic Data","created":"2025-10-08","cited_by_count":4,"area":"alignment"},{"id":"2603.24124","title":"The Alignment Tax: Response Homogenization in Aligned LLMs and Its Implications for Uncertainty Estimation","created":"2026-03-25","cited_by_count":4,"area":"alignment"},{"id":"2502.00136","title":"A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment","created":"2025-01-31","cited_by_count":3,"area":"alignment"},{"id":"2502.08657","title":"Refining Positive and Toxic Samples for Dual Safety Self-Alignment of LLMs with Minimal Human Interventions","created":"2025-02-08","cited_by_count":3,"area":"alignment"},{"id":"2503.23777","title":"CONGRAD:Conflicting Gradient Filtering for Multilingual Preference Alignment","created":"2025-03-31","cited_by_count":3,"area":"alignment"},{"id":"2505.19327","title":"Paying Alignment Tax with Contrastive Learning","created":"2025-05-25","cited_by_count":3,"area":"alignment"},{"id":"2506.15717","title":"daDPO: Distribution-Aware DPO for Distilling Conversational Abilities","created":"2025-06-03","cited_by_count":3,"area":"alignment"},{"id":"2602.06763","title":"R-Align: Enhancing Generative Reward Models through Rationale-Centric Meta-Judging","created":"2026-02-06","cited_by_count":3,"area":"alignment"},{"id":"2603.17044","title":"Do Understanding and Generation Fight? A Diagnostic Study of DPO for Unified Multimodal Models","created":"2026-03-17","cited_by_count":3,"area":"alignment"},{"id":"2606.05523","title":"CHASE: Adversarial Red-Blue Teaming for Improving LLM Safety using Reinforcement Learning","created":"2026-06-04","cited_by_count":3,"area":"alignment"},{"id":"2409.18417","title":"VickreyFeedback: Cost-efficient Data Construction for Reinforcement Learning from Human Feedback","created":"2024-09-27","cited_by_count":2,"area":"alignment"},{"id":"2410.00508","title":"FlipGuard: Defending Preference Alignment against Update Regression with Constrained Optimization","created":"2024-10-01","cited_by_count":2,"area":"alignment"},{"id":"2412.00074","title":"Safe to Serve: Aligning Instruction-Tuned Models for Safety and Helpfulness","created":"2024-11-26","cited_by_count":2,"area":"alignment"},{"id":"2510.03269","title":"General Exploratory Bonus for Optimistic Exploration in RLHF","created":"2025-09-27","cited_by_count":2,"area":"alignment"},{"id":"2601.05882","title":"An Empirical Study on Preference Tuning Generalization and Diversity Under Domain Shift","created":"2026-01-09","cited_by_count":2,"area":"alignment"},{"id":"2602.11661","title":"Quark Medical Alignment: A Holistic Multi-Dimensional Alignment and Collaborative Optimization Paradigm","created":"2026-02-12","cited_by_count":2,"area":"alignment"},{"id":"2606.21906","title":"Deeper is Not Always Better: Mitigating the Alignment Tax via Confident Layer Decoding","created":"2026-06-20","cited_by_count":2,"area":"alignment"},{"id":"2505.13257","title":"Is Active Persona Inference Necessary for Aligning Small Models to Personal Preferences?","created":"2025-05-19","cited_by_count":1,"area":"alignment"},{"id":"2512.03704","title":"Overcoming State Inertia: Minimally Invasive Temporal Alignment for Evolving Contexts","created":"2025-12-03","cited_by_count":1,"area":"alignment"},{"id":"2604.00979","title":"Dual Optimal: Make Your LLM Peer-like with Dignity","created":"2026-04-01","cited_by_count":1,"area":"alignment"},{"id":"2605.11679","title":"Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion","created":"2026-05-12","cited_by_count":1,"area":"alignment"},{"id":"2606.21089","title":"Repeated post-training is not Self-improving: Diagnosing Scientific Amnesia in Continual DPO Pipelines","created":"2026-06-17","cited_by_count":1,"area":"alignment"},{"id":"2607.16240","title":"Normalized Rewards for Preference Optimization","created":"2026-06-26","cited_by_count":1,"area":"alignment"},{"id":"2607.29246","title":"Don't Mix Rewards, Mix Policies: Policy Decomposition and Optimization for Multi-Reward RL","created":"2026-07-31","cited_by_count":1,"area":"alignment"},{"id":"2506.03557","title":"BPO: Revisiting Preference Modeling in Direct Preference Optimization","created":"2025-06-04","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":"2601.09718","title":"StatLLaMA: Multi-Stage training for domain-optimized statistical large language models","created":"2025-12-26","cited_by_count":0,"area":"alignment"},{"id":"2602.04909","title":"Learning Where It Matters: Geometric Anchoring for Robust Preference Alignment","created":"2026-02-04","cited_by_count":0,"area":"alignment"},{"id":"2603.04822","title":"VISA: Value Injection via Shielded Adaptation for Personalized LLM Alignment","created":"2026-03-05","cited_by_count":0,"area":"alignment"},{"id":"2604.06833","title":"FedDetox: Robust Federated SLM Alignment via On-Device Data Sanitization","created":"2026-04-08","cited_by_count":0,"area":"alignment"},{"id":"2604.20685","title":"MGDA-Decoupled: Geometry-Aware Multi-Objective Optimisation for DPO-based LLM Alignment","created":"2026-04-22","cited_by_count":0,"area":"alignment"},{"id":"2607.03453","title":"Best-of-Better-$N$: Generating Pre-Aligned Responses with In-Context Learning","created":"2026-07-03","cited_by_count":0,"area":"alignment"},{"id":"2608.06179","title":"SAGA: Score-Weighted Adaptive Generation Alignment for Low-Resource Nordic Language Models","created":"2026-08-06","cited_by_count":0,"area":"alignment"},{"id":"2608.22490","title":"Who Pays More for Safety? Measuring the Disparate Cost of Safety Alignment across Languages","created":"2026-08-23","cited_by_count":0,"area":"alignment"},{"id":"2608.24192","title":"Preference Data Selection for Mitigating the Alignment Tax in Large Language Models","created":"2026-08-25","cited_by_count":0,"area":"alignment"},{"id":"2609.33221","title":"RMB: Reward Model Boosting Mitigates Reward Hacking","created":"2026-09-27","cited_by_count":0,"area":"alignment"}],"next_cursor":null}