{"total":120,"rows":[{"id":"2305.18290","title":"Direct Preference Optimization: Your Language Model is Secretly a Reward Model","created":"2023-05-29","cited_by_count":8065,"area":"lr"},{"id":"2307.15217","title":"Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback","created":"2023-07-27","cited_by_count":759,"area":"alignment"},{"id":"2310.06987","title":"Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation","created":"2023-10-10","cited_by_count":451,"area":"safety"},{"id":"2402.01306","title":"KTO: Model Alignment as Prospect Theoretic Optimization","created":"2024-02-02","cited_by_count":264,"area":"lr"},{"id":"2306.04488","title":"Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards","created":"2023-06-07","cited_by_count":243,"area":"alignment"},{"id":"2309.15025","title":"Large Language Model Alignment: A Survey","created":"2023-09-26","cited_by_count":243,"area":"alignment"},{"id":"2308.03188","title":"Automatically Correcting Large Language Models: Surveying the landscape of diverse self-correction strategies","created":"2023-08-06","cited_by_count":238,"area":"lr"},{"id":"2309.14348","title":"Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM","created":"2023-09-18","cited_by_count":192,"area":"alignment"},{"id":"2309.16240","title":"Beyond Reverse KL: Generalizing Direct Preference Optimization with Diverse Divergence Constraints","created":"2023-09-28","cited_by_count":174,"area":"alignment"},{"id":"2402.09739","title":"QuRating: Selecting High-Quality Data for Training Language Models","created":"2024-02-15","cited_by_count":157,"area":null},{"id":"2311.12786","title":"Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks","created":"2023-11-21","cited_by_count":115,"area":null},{"id":"2504.03380","title":"Online Difficulty Filtering for Reasoning Oriented Reinforcement Learning","created":"2025-04-04","cited_by_count":104,"area":"lr"},{"id":"2401.13136","title":"The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts","created":"2024-01-23","cited_by_count":94,"area":"alignment"},{"id":"2504.05520","title":"Efficient Reinforcement Finetuning via Adaptive Curriculum Learning","created":"2025-04-07","cited_by_count":93,"area":"lr"},{"id":"2404.00929","title":"A Survey on Multilingual Large Language Models: Corpora, Alignment, and Bias","created":"2024-04-01","cited_by_count":84,"area":null},{"id":"2410.07471","title":"SEAL: Safety-enhanced Aligned LLM Fine-tuning via Bilevel Data Selection","created":"2024-10-09","cited_by_count":64,"area":null},{"id":"2401.00243","title":"Uncertainty-Penalized Reinforcement Learning from Human Feedback with Diverse Reward LoRA Ensembles","created":"2023-12-30","cited_by_count":56,"area":"alignment"},{"id":"2403.03185","title":"Correlated Proxies: A New Definition and Improved Mitigation for Reward Hacking","created":"2024-03-05","cited_by_count":50,"area":"alignment"},{"id":"2407.21035","title":"Direct Unlearning Optimization for Robust and Safe Text-to-Image Models","created":"2024-07-17","cited_by_count":50,"area":"safety"},{"id":"2306.11816","title":"Learning to Generate Better Than Your LLM","created":"2023-06-20","cited_by_count":49,"area":null},{"id":"2310.08184","title":"Learning from models beyond fine-tuning","created":"2023-10-12","cited_by_count":48,"area":null},{"id":"2405.19534","title":"Preference Learning Algorithms Do Not Learn Preference Rankings","created":"2024-05-29","cited_by_count":48,"area":"alignment"},{"id":"2503.17126","title":"Modifying Large Language Model Post-Training for Diverse Creative Writing","created":"2025-03-21","cited_by_count":40,"area":"alignment"},{"id":"2312.00027","title":"Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections","created":"2023-11-15","cited_by_count":39,"area":"safety"},{"id":"2406.16306","title":"Cascade Reward Sampling for Efficient Decoding-Time Alignment","created":"2024-06-24","cited_by_count":39,"area":"alignment"},{"id":"2407.09499","title":"Self-Consuming Generative Models with Curated Data Provably Optimize Human Preferences","created":"2024-06-12","cited_by_count":39,"area":"alignment"},{"id":"2406.17692","title":"From Distributional to Overton Pluralism: Investigating Large Language Model Alignment","created":"2024-06-25","cited_by_count":33,"area":"alignment"},{"id":"2410.06101","title":"Coevolving with the Other You: Fine-Tuning LLM with Sequential Cooperative Multi-Agent Reinforcement Learning","created":"2024-10-08","cited_by_count":33,"area":"agents"},{"id":"2306.00398","title":"Preference-grounded Token-level Guidance for Language Model Fine-tuning","created":"2023-06-01","cited_by_count":31,"area":"alignment"},{"id":"2405.10313","title":"How Far Are We From AGI: Are LLMs All We Need?","created":"2024-05-16","cited_by_count":31,"area":null},{"id":"2510.20817","title":"KL-Regularized Reinforcement Learning is Designed to Mode Collapse","created":"2025-10-23","cited_by_count":31,"area":"deeprl"},{"id":"2309.16155","title":"The Trickle-down Impact of Reward (In-)consistency on RLHF","created":"2023-09-28","cited_by_count":29,"area":"alignment"},{"id":"2402.10693","title":"Exploring Precision and Recall to assess the quality and diversity of LLMs","created":"2024-02-16","cited_by_count":29,"area":null},{"id":"2405.10529","title":"Safeguarding Vision-Language Models Against Patched Visual Prompt Injectors","created":"2024-05-17","cited_by_count":27,"area":"mllm"},{"id":"2406.04412","title":"Spread Preference Annotation: Direct Preference Judgment for Efficient LLM Alignment","created":"2024-06-06","cited_by_count":27,"area":"alignment"},{"id":"2310.13011","title":"Compositional preference models for aligning LMs","created":"2023-10-17","cited_by_count":26,"area":"alignment"},{"id":"2312.15907","title":"Align on the Fly: Adapting Chatbot Behavior to Established Norms","created":"2023-12-26","cited_by_count":23,"area":"alignment"},{"id":"2411.03817","title":"From Novice to Expert: LLM Agent Policy Optimization via Step-wise Reinforcement Learning","created":"2024-11-06","cited_by_count":22,"area":"agents"},{"id":"2403.04204","title":"On the Essence and Prospect: An Investigation of Alignment Approaches for Big Models","created":"2024-03-07","cited_by_count":21,"area":"alignment"},{"id":"2309.09055","title":"Exploring the impact of low-rank adaptation on the performance, efficiency, and regularization of RLHF","created":"2023-09-16","cited_by_count":20,"area":"alignment"},{"id":"2407.02446","title":"Predicting vs. Acting: A Trade-off Between World Modeling & Agent Modeling","created":"2024-07-02","cited_by_count":20,"area":"wm"},{"id":"2305.14718","title":"Leftover Lunch: Advantage-based Offline Reinforcement Learning for Language Models","created":"2023-05-24","cited_by_count":17,"area":"alignment"},{"id":"2311.01070","title":"Multilingual DistilWhisper: Efficient Distillation of Multi-task Speech Models via Language-Specific Experts","created":"2023-11-02","cited_by_count":17,"area":null},{"id":"2406.00049","title":"QUEST: Quality-Aware Metropolis-Hastings Sampling for Machine Translation","created":"2024-05-28","cited_by_count":16,"area":null},{"id":"2410.12138","title":"Preference Optimization with Multi-Sample Comparisons","created":"2024-10-16","cited_by_count":16,"area":"diffusion"},{"id":"2510.04474","title":"DRPO: Efficient Reasoning via Decoupled Reward Policy Optimization","created":"2025-10-06","cited_by_count":16,"area":"lr"},{"id":"2409.09774","title":"Generalizing Alignment Paradigm of Text-to-Image Generation with Preferences through $f$-divergence Minimization","created":"2024-09-15","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":"2410.05328","title":"Reward Learning From Preference With Ties","created":"2024-10-05","cited_by_count":14,"area":"alignment"},{"id":"2502.11026","title":"RLHF in an SFT Way: From Optimal Solution to Reward-Weighted Alignment","created":"2025-02-16","cited_by_count":14,"area":"alignment"},{"id":"2503.21720","title":"Collab: Controlled Decoding using Mixture of Agents for LLM Alignment","created":"2025-03-27","cited_by_count":14,"area":"agents"},{"id":"2501.02790","title":"Segmenting Text and Learning Their Rewards for Improved RLHF in Language Model","created":"2025-01-06","cited_by_count":12,"area":"alignment"},{"id":"2308.14328","title":"Reinforcement Learning for Generative AI: A Survey","created":"2023-08-28","cited_by_count":11,"area":null},{"id":"2403.03419","title":"Negating Negatives: Alignment with Human Negative Samples via Distributional Dispreference Optimization","created":"2024-03-06","cited_by_count":11,"area":"alignment"},{"id":"2410.02596","title":"Beyond Squared Error: Exploring Loss Design for Enhanced Training of Generative Flow Networks","created":"2024-10-03","cited_by_count":11,"area":null},{"id":"2507.00445","title":"Iterative Distillation for Reward-Guided Fine-Tuning of Diffusion Models in Biomolecular Design","created":"2025-07-01","cited_by_count":11,"area":"diffusion"},{"id":"2411.08302","title":"RED: Unleashing Token-Level Rewards from Holistic Feedback via Reward Redistribution","created":"2024-11-13","cited_by_count":10,"area":"alignment"},{"id":"2501.18537","title":"Loss Functions and Operators Generated by f-Divergences","created":"2025-01-30","cited_by_count":10,"area":null},{"id":"2505.19428","title":"Frictional Agent Alignment Framework: Slow Down and Don't Break Things","created":"2025-05-26","cited_by_count":10,"area":"agents"},{"id":"2510.01243","title":"Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editing","created":"2025-09-24","cited_by_count":10,"area":"alignment"},{"id":"2310.17551","title":"Unpacking the Ethical Value Alignment in Big Models","created":"2023-10-26","cited_by_count":9,"area":null},{"id":"2404.04920","title":"Regularized Conditional Diffusion Model for Multi-Task Preference Alignment","created":"2024-04-07","cited_by_count":9,"area":"diffusion"},{"id":"2408.15625","title":"CBF-LLM: Safe Control for LLM Alignment","created":"2024-08-28","cited_by_count":9,"area":"alignment"},{"id":"2505.18384","title":"Dynamic Risk Assessments for Offensive Cybersecurity Agents","created":"2025-05-23","cited_by_count":8,"area":"agents"},{"id":"2512.15605","title":"Autoregressive Language Models are Secretly Energy-Based Models: Insights into the Lookahead Capabilities of Next-Token Prediction","created":"2025-12-17","cited_by_count":8,"area":null},{"id":"2405.12900","title":"Adversarial DPO: Harnessing Harmful Data for Reducing Toxicity with Minimal Impact on Coherence and Evasiveness in Dialogue Agents","created":"2024-05-21","cited_by_count":7,"area":"agents"},{"id":"2410.06716","title":"Guaranteed Generation from Large Language Models","created":"2024-10-09","cited_by_count":7,"area":null},{"id":"2505.04993","title":"Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes","created":"2025-05-08","cited_by_count":7,"area":"alignment"},{"id":"2506.06376","title":"Enhancing Decision-Making of Large Language Models via Actor-Critic","created":"2025-06-04","cited_by_count":7,"area":null},{"id":"2508.09654","title":"Improving Diversity in Language Models: When Temperature Fails, Change the Loss","created":"2025-08-13","cited_by_count":7,"area":null},{"id":"2605.15113","title":"Learning from Language Feedback via Variational Policy Distillation","created":"2026-05-14","cited_by_count":7,"area":"distill"},{"id":"2410.20187","title":"Uncertainty-Penalized Direct Preference Optimization","created":"2024-10-26","cited_by_count":6,"area":"alignment"},{"id":"2510.10020","title":"Calibrating Generative Models to Distributional Constraints","created":"2025-10-11","cited_by_count":6,"area":null},{"id":"2406.07971","title":"It Takes Two: On the Seamlessness between Reward and Policy Model in RLHF","created":"2024-06-12","cited_by_count":5,"area":"alignment"},{"id":"2409.02718","title":"\"Yes, My LoRD.\" Guiding Language Model Extraction with Locality Reinforced Distillation","created":"2024-09-04","cited_by_count":5,"area":null},{"id":"2503.00030","title":"RSPO: Regularized Self-Play Alignment of Large Language Models","created":"2025-02-24","cited_by_count":5,"area":"alignment"},{"id":"2506.09656","title":"Multi-level Value Alignment in Agentic AI Systems: Survey and Perspectives","created":"2025-06-11","cited_by_count":5,"area":"agents"},{"id":"2602.23116","title":"Provably Efficient Regularized Online RLHF with Generalized Bilinear Preferences","created":"2026-02-26","cited_by_count":5,"area":"alignment"},{"id":"2603.14128","title":"Diffusion Reinforcement Learning via Centered Reward Distillation","created":"2026-03-14","cited_by_count":5,"area":"diffusion"},{"id":"2503.10093","title":"Efficient Safety Alignment of Large Language Models via Preference Re-ranking and Representation-based Reward Modeling","created":"2025-03-13","cited_by_count":4,"area":"alignment"},{"id":"2508.15044","title":"Reward-Shifted Speculative Sampling Is An Efficient Test-Time Weak-to-Strong Aligner","created":"2025-08-20","cited_by_count":4,"area":"alignment"},{"id":"2509.20368","title":"LATTS: Locally Adaptive Test-Time Scaling","created":"2025-09-16","cited_by_count":4,"area":"lr"},{"id":"2509.21884","title":"You Can't Steal Nothing: Mitigating Prompt Leakages in LLMs via System Vectors","created":"2025-09-26","cited_by_count":4,"area":null},{"id":"2605.15417","title":"$f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data","created":"2026-05-14","cited_by_count":4,"area":null},{"id":"2303.05431","title":"disco: a toolkit for Distributional Control of Generative Models","created":"2023-03-08","cited_by_count":3,"area":null},{"id":"2501.13726","title":"RPO: Retrieval Preference Optimization for Robust Retrieval-Augmented Generation","created":"2025-01-23","cited_by_count":3,"area":"rag"},{"id":"2504.10002","title":"FLoRA: Sample-Efficient Preference-based RL via Low-Rank Style Adaptation of Reward Functions","created":"2025-04-14","cited_by_count":3,"area":"alignment"},{"id":"2505.21907","title":"Modeling and Optimizing User Preferences in AI Copilots: A Comprehensive Survey and Taxonomy","created":"2025-05-28","cited_by_count":3,"area":"alignment"},{"id":"2510.18353","title":"Ranking-based Preference Optimization for Diffusion Models from Implicit User Feedback","created":"2025-10-21","cited_by_count":3,"area":"diffusion"},{"id":"2512.00709","title":"When Human Preferences Flip: An Instance-Dependent Robust Loss for RLHF","created":"2025-11-30","cited_by_count":3,"area":"alignment"},{"id":"2512.05962","title":"Whatever Remains Must Be True: Filtering Drives Reasoning in LLMs, Shaping Diversity","created":"2025-12-05","cited_by_count":3,"area":"lr"},{"id":"2310.15694","title":"COPR: Continual Learning Human Preference through Optimal Policy Regularization","created":"2023-10-24","cited_by_count":2,"area":"alignment"},{"id":"2405.00966","title":"Efficient Compression of Multitask Multilingual Speech Models","created":"2024-05-02","cited_by_count":2,"area":null},{"id":"2501.01741","title":"How Toxic Can You Get? Search-based Toxicity Testing for Large Language Models","created":"2025-01-03","cited_by_count":2,"area":null},{"id":"2512.16565","title":"Non-Asymptotic Global Convergence of PPO-Clip","created":"2025-12-18","cited_by_count":2,"area":null},{"id":"2605.04653","title":"Threshold-Guided Optimization for Visual Generative Models","created":"2026-05-06","cited_by_count":2,"area":"alignment"},{"id":"2605.06977","title":"$f$-Divergence Regularized RLHF: Two Tales of Sampling and Unified Analyses","created":"2026-05-07","cited_by_count":2,"area":"alignment"},{"id":"2606.12360","title":"Anatomy of Post-Training: Using Interpretability to Characterize Data and Shape the Learning Signal","created":"2026-06-10","cited_by_count":2,"area":null},{"id":"2404.12150","title":"Aligning language models with human preferences","created":"2024-04-18","cited_by_count":1,"area":"alignment"},{"id":"2501.04568","title":"Feedback-Driven Vision-Language Alignment with Minimal Human Supervision","created":"2025-01-08","cited_by_count":1,"area":"mllm"}],"next_cursor":"100"}