{"total":126,"rows":[{"id":"2506.01937","title":"RewardBench 2: Advancing Reward Model Evaluation","created":"2025-06-02","cited_by_count":150,"area":"alignment"},{"id":"2510.07743","title":"OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignment","created":"2025-10-09","cited_by_count":132,"area":"alignment"},{"id":"2507.18624","title":"Checklists Are Better Than Reward Models For Aligning Language Models","created":"2025-07-24","cited_by_count":128,"area":"alignment"},{"id":"2410.14872","title":"How to Evaluate Reward Models for RLHF","created":"2024-10-18","cited_by_count":88,"area":"alignment"},{"id":"2402.16192","title":"Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing","created":"2024-02-25","cited_by_count":74,"area":"alignment"},{"id":"2406.13542","title":"Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models","created":"2024-06-19","cited_by_count":73,"area":"alignment"},{"id":"2504.12501","title":"Reinforcement Learning from Human Feedback","created":"2025-04-16","cited_by_count":58,"area":"alignment"},{"id":"2412.14516","title":"Cal-DPO: Calibrated Direct Preference Optimization for Language Model Alignment","created":"2024-12-19","cited_by_count":51,"area":"alignment"},{"id":"2407.00617","title":"Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning","created":"2024-06-30","cited_by_count":45,"area":"alignment"},{"id":"2409.20370","title":"The Perfect Blend: Redefining RLHF with Mixture of Judges","created":"2024-09-30","cited_by_count":43,"area":"alignment"},{"id":"2510.05024","title":"Inoculation Prompting: Instructing LLMs to misbehave at train-time improves test-time alignment","created":"2025-10-06","cited_by_count":43,"area":"alignment"},{"id":"2412.15838","title":"Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback","created":"2024-12-20","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":"2503.11701","title":"A Survey of Direct Preference Optimization","created":"2025-03-12","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":"2409.17407","title":"Post-hoc Reward Calibration: A Case Study on Length Bias","created":"2024-09-25","cited_by_count":29,"area":"alignment"},{"id":"2603.12273","title":"Aligning Language Models from User Interactions","created":"2026-02-18","cited_by_count":28,"area":"alignment"},{"id":"2410.19133","title":"Hybrid Preferences: Learning to Route Instances for Human vs. AI Feedback","created":"2024-10-24","cited_by_count":26,"area":"alignment"},{"id":"2502.00883","title":"SimPER: A Minimalist Approach to Preference Alignment without Hyperparameters","created":"2025-02-02","cited_by_count":25,"area":"alignment"},{"id":"2504.14716","title":"Pairwise or Pointwise? Evaluating Feedback Protocols for Bias in LLM-Based Evaluation","created":"2025-04-20","cited_by_count":24,"area":"alignment"},{"id":"2504.04950","title":"A Unified Pairwise Framework for RLHF: Bridging Generative Reward Modeling and Policy Optimization","created":"2025-04-07","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":"2509.01909","title":"Oyster-I: Beyond Refusal -- Constructive Safety Alignment for Responsible Language Models","created":"2025-09-02","cited_by_count":19,"area":"alignment"},{"id":"2605.09959","title":"G-Zero: Self-Play for Open-Ended Generation from Zero Data","created":"2026-05-11","cited_by_count":17,"area":"alignment"},{"id":"2505.20259","title":"Lifelong Safety Alignment for Language Models","created":"2025-05-26","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":"2505.19653","title":"Token-Importance Guided Direct Preference Optimization","created":"2025-05-26","cited_by_count":14,"area":"alignment"},{"id":"2412.09544","title":"Sail into the Headwind: Alignment via Robust Rewards and Dynamic Labels against Reward Hacking","created":"2024-12-12","cited_by_count":13,"area":"alignment"},{"id":"2505.10527","title":"WorldPM: Scaling Human Preference Modeling","created":"2025-05-15","cited_by_count":13,"area":"alignment"},{"id":"2605.07396","title":"Rubric-based On-policy Distillation","created":"2026-05-08","cited_by_count":13,"area":"alignment"},{"id":"2410.03717","title":"Revisiting the Superficial Alignment Hypothesis","created":"2024-09-27","cited_by_count":12,"area":"alignment"},{"id":"2412.01253","title":"Yi-Lightning Technical Report","created":"2024-12-02","cited_by_count":12,"area":"alignment"},{"id":"2505.11166","title":"SoLoPO: Unlocking Long-Context Capabilities in LLMs via Short-to-Long Preference Optimization","created":"2025-05-16","cited_by_count":10,"area":"alignment"},{"id":"2512.16626","title":"Stackelberg Learning from Human Feedback: Preference Optimization as a Sequential Game","created":"2025-12-18","cited_by_count":10,"area":"alignment"},{"id":"2410.21438","title":"UFT: Unifying Fine-Tuning of SFT and RLHF/DPO/UNA through a Generalized Implicit Reward Function","created":"2024-10-28","cited_by_count":9,"area":"alignment"},{"id":"2506.10054","title":"Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs","created":"2025-06-11","cited_by_count":9,"area":"alignment"},{"id":"2507.21983","title":"Improving Generative Ad Text on Facebook using Reinforcement Learning","created":"2025-07-29","cited_by_count":9,"area":"alignment"},{"id":"2412.15282","title":"A Systematic Examination of Preference Learning through the Lens of Instruction-Following","created":"2024-12-18","cited_by_count":8,"area":"alignment"},{"id":"2505.12762","title":"IDEAL: Data Equilibrium Adaptation for Multi-Capability Language Model Alignment","created":"2025-05-19","cited_by_count":8,"area":"alignment"},{"id":"2505.13878","title":"InfiFPO: Implicit Model Fusion via Preference Optimization in Large Language Models","created":"2025-05-20","cited_by_count":8,"area":"alignment"},{"id":"2512.23457","title":"Replay Failures as Successes: Sample-Efficient Reinforcement Learning for Instruction Following","created":"2025-12-29","cited_by_count":8,"area":"alignment"},{"id":"2602.04896","title":"Steering Externalities: Benign Activation Steering Unintentionally Increases Jailbreak Risk for Large Language Models","created":"2026-02-03","cited_by_count":8,"area":"alignment"},{"id":"2405.20830","title":"Self-Augmented Preference Optimization: Off-Policy Paradigms for Language Model Alignment","created":"2024-05-31","cited_by_count":7,"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":"2507.13158","title":"Inverse Reinforcement Learning Meets Large Language Model Post-Training: Basics, Advances, and Opportunities","created":"2025-07-17","cited_by_count":7,"area":"alignment"},{"id":"2509.23102","title":"Multiplayer Nash Preference Optimization","created":"2025-09-27","cited_by_count":7,"area":"alignment"},{"id":"2510.05052","title":"Proactive defense against LLM Jailbreak","created":"2025-10-06","cited_by_count":7,"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":"2602.06258","title":"GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt","created":"2026-02-05","cited_by_count":7,"area":"alignment"},{"id":"2408.15339","title":"UNA: A Unified Supervised Framework for Efficient LLM Alignment Across Feedback Types","created":"2024-08-27","cited_by_count":6,"area":"alignment"},{"id":"2410.14940","title":"Baichuan Alignment Technical Report","created":"2024-10-19","cited_by_count":6,"area":"alignment"},{"id":"2505.02363","title":"SIMPLEMIX: Frustratingly Simple Mixing of Off- and On-policy Data in Language Model Preference Learning","created":"2025-05-05","cited_by_count":6,"area":"alignment"},{"id":"2510.08697","title":"BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution","created":"2025-10-09","cited_by_count":6,"area":"alignment"},{"id":"2510.24235","title":"PaTaRM: Bridging Pairwise and Pointwise Signals via Preference-Aware Task-Adaptive Reward Modeling","created":"2025-10-28","cited_by_count":6,"area":"alignment"},{"id":"2601.18533","title":"From Verifiable Dot to Reward Chain: Harnessing Verifiable Reference-based Rewards for Reinforcement Learning of Open-ended Generation","created":"2026-01-26","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":"2509.22099","title":"S2J: Bridging the Gap Between Solving and Judging Ability in Generative Reward Models","created":"2025-09-26","cited_by_count":5,"area":"alignment"},{"id":"2602.03876","title":"GOPO: Policy Optimization using Ranked Rewards","created":"2026-02-01","cited_by_count":5,"area":"alignment"},{"id":"2602.10623","title":"Mitigating Reward Hacking in RLHF via Bayesian Non-negative Reward Modeling","created":"2026-02-11","cited_by_count":5,"area":"alignment"},{"id":"2602.15823","title":"CrispEdit: Low-Curvature Projections for Scalable Non-Destructive LLM Editing","created":"2026-02-17","cited_by_count":5,"area":"alignment"},{"id":"2411.07595","title":"Entropy Controllable Direct Preference Optimization","created":"2024-11-12","cited_by_count":4,"area":"alignment"},{"id":"2502.11284","title":"Balancing the Budget: Understanding Trade-offs Between Supervised and Preference-Based Finetuning","created":"2025-02-16","cited_by_count":4,"area":"alignment"},{"id":"2602.00986","title":"Sparse Reward Subsystem in Large Language Models","created":"2026-02-01","cited_by_count":4,"area":"alignment"},{"id":"2506.00835","title":"SynPO: Synergizing Descriptiveness and Preference Optimization for Video Detailed Captioning","created":"2025-06-01","cited_by_count":3,"area":"alignment"},{"id":"2511.20604","title":"On Evaluating LLM Alignment by Evaluating LLMs as Judges","created":"2025-11-25","cited_by_count":3,"area":"alignment"},{"id":"2601.22801","title":"Clipping-Free Policy Optimization for Large Language Models","created":"2026-01-30","cited_by_count":3,"area":"alignment"},{"id":"2605.18592","title":"AMARIS: A Memory-Augmented Rubric Improvement System for Rubric-Based Reinforcement Learning","created":"2026-05-18","cited_by_count":3,"area":"alignment"},{"id":"2606.21795","title":"Discretizing Reward Models","created":"2026-06-19","cited_by_count":3,"area":"alignment"},{"id":"2501.03271","title":"DPO Kernels: A Semantically-Aware, Kernel-Enhanced, and Divergence-Rich Paradigm for Direct Preference Optimization","created":"2025-01-05","cited_by_count":2,"area":"alignment"},{"id":"2504.08590","title":"Playpen: An Environment for Exploring Learning Through Conversational Interaction","created":"2025-04-11","cited_by_count":2,"area":"alignment"},{"id":"2506.04272","title":"Understanding the Impact of Sampling Quality in Direct Preference Optimization","created":"2025-06-03","cited_by_count":2,"area":"alignment"},{"id":"2508.08243","title":"Jinx: Unlimited LLMs for Probing Alignment Failures","created":"2025-08-11","cited_by_count":2,"area":"alignment"},{"id":"2510.07315","title":"SWE-IF: Aligning Code Evaluation with Human Preference","created":"2025-10-08","cited_by_count":2,"area":"alignment"},{"id":"2602.06869","title":"Uncovering Cross-Objective Interference in Multi-Objective Alignment","created":"2026-02-06","cited_by_count":2,"area":"alignment"},{"id":"2603.09692","title":"ActiveUltraFeedback: Efficient Preference Data Generation using Active Learning","created":"2026-03-10","cited_by_count":2,"area":"alignment"},{"id":"2603.19311","title":"PrefPO: Pairwise Preference Prompt Optimization","created":"2026-03-13","cited_by_count":2,"area":"alignment"},{"id":"2606.16276","title":"SpecAlign: Efficient Specification-Grounded Alignment of Large Language Models via Synthetic Data","created":"2026-06-15","cited_by_count":2,"area":"alignment"},{"id":"2502.18679","title":"Discriminative Finetuning of Generative Large Language Models without Reward Models and Human Preference Data","created":"2025-02-25","cited_by_count":1,"area":"alignment"},{"id":"2506.09457","title":"Towards Bridging the Reward-Generation Gap in Direct Alignment Algorithms","created":"2025-06-11","cited_by_count":1,"area":"alignment"},{"id":"2506.13702","title":"Value-Free Policy Optimization via Reward Partitioning","created":"2025-06-16","cited_by_count":1,"area":"alignment"},{"id":"2509.25148","title":"AAPA: Adversarially Anchored Preference Alignment for Post-Training of Large Language Models","created":"2025-09-29","cited_by_count":1,"area":"alignment"},{"id":"2510.21090","title":"Self-Rewarding PPO: Aligning Large Language Models with Demonstrations Only","created":"2025-10-24","cited_by_count":1,"area":"alignment"},{"id":"2511.10985","title":"When Data is the Algorithm: A Systematic Study and Curation of Preference Optimization Datasets","created":"2025-11-14","cited_by_count":1,"area":"alignment"},{"id":"2512.17920","title":"Separating Constraint Compliance from Semantic Accuracy: A Novel Benchmark for Evaluating Instruction-Following Under Compression","created":"2025-12-02","cited_by_count":1,"area":"alignment"},{"id":"2601.22083","title":"Latent Adversarial Regularization for Offline Preference Optimization","created":"2026-01-29","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":"2603.02225","title":"Scaling Reward Modeling without Human Supervision","created":"2026-02-11","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":"2605.20285","title":"Introspective X Training: Feedback Conditioning Improves Scaling Across all LLM Training Stages","created":"2026-05-19","cited_by_count":1,"area":"alignment"},{"id":"2605.20834","title":"Conditional Equivalence of DPO and RLHF: Implicit Assumption, Failure Modes, and Provable Alignment","created":"2026-05-20","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":"2609.14896","title":"Forty Shades of Blue: Quality-Diversity Alignment via Mode-Conditioned Reinforcement Learning","created":"2026-09-14","cited_by_count":1,"area":"alignment"},{"id":"2505.12845","title":"Multi-Level Aware Preference Learning: Enhancing RLHF for Complex Multi-Instruction Tasks","created":"2025-05-19","cited_by_count":0,"area":"alignment"},{"id":"2506.02018","title":"Enhancing Paraphrase Type Generation: The Impact of DPO and RLHF Evaluated with Human-Ranked Data","created":"2025-05-28","cited_by_count":0,"area":"alignment"},{"id":"2510.09887","title":"Overconfident and Blind to Details: Fixing Prompt Insensitivity with Abductive Preference Learning","created":"2025-10-10","cited_by_count":0,"area":"alignment"},{"id":"2510.15746","title":"LLMs Judge Themselves: A Game-Theoretic Framework for Human-Aligned Evaluation","created":"2025-10-17","cited_by_count":0,"area":"alignment"},{"id":"2601.06157","title":"ECLIPTICA -- A Framework for Switchable LLM Alignment via CITA - Contrastive Instruction-Tuned Alignment","created":"2026-01-06","cited_by_count":0,"area":"alignment"},{"id":"2602.02898","title":"Aligning Language Model Benchmarks with Pairwise Preferences","created":"2026-02-02","cited_by_count":0,"area":"alignment"},{"id":"2603.18637","title":"MOSAIC: Multi-Objective Slice-Aware Iterative Curation for Alignment","created":"2026-03-19","cited_by_count":0,"area":"alignment"},{"id":"2604.02574","title":"Understanding the Effects of Safety Unalignment on Large Language Models","created":"2026-04-02","cited_by_count":0,"area":"alignment"}],"next_cursor":"100"}