English
Related papers

Related papers: RSPO: Regularized Self-Play Alignment of Large Lan…

200 papers

Preference optimization methods typically begin training with a well-trained SFT model as a reference model. In RLHF and DPO, a regularization term is used during the preference optimization process to prevent the policy model from…

Computation and Language · Computer Science 2024-09-30 Cheolhun Jang

Supervised fine-tuning (SFT) has emerged as a crucial method for aligning large language models (LLMs) with human-annotated demonstrations. However, SFT, being an off-policy approach similar to behavior cloning, often struggles with…

Computation and Language · Computer Science 2025-10-27 Qingru Zhang , Liang Qiu , Ilgee Hong , Zhenghao Xu , Tianyi Liu , Shiyang Li , Rongzhi Zhang , Zheng Li , Lihong Li , Bing Yin , Chao Zhang , Jianshu Chen , Haoming Jiang , Tuo Zhao

While Reinforcement Learning (RL) shows promise in training tool-use Large Language Models (LLMs) using verifiable outcome rewards, existing methods largely overlook the potential of reasoning rewards based on chain-of-thought quality for…

Computation and Language · Computer Science 2026-01-16 Zihan Lin , Xiaohan Wang , Hexiong Yang , Jiajun Chai , Jie Cao , Guojun Yin , Wei Lin , Ran He

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a promising approach to improve the reasoning abilities of Large Language Models (LLMs). Among RLVR algorithms, Group Relative Policy Optimization (GRPO) and its variants…

Artificial Intelligence · Computer Science 2026-04-21 Zhaokang Liao , Yingguo Gao , Yi Yang , Yongheng Hu , Jingting Ding

Aligning Large Language Models (LLMs) traditionally relies on costly training and human preference annotations. Self-alignment seeks to reduce these expenses by enabling models to align themselves. To further lower costs and achieve…

Computation and Language · Computer Science 2024-11-15 Somanshu Singla , Zhen Wang , Tianyang Liu , Abdullah Ashfaq , Zhiting Hu , Eric P. Xing

Preference alignment is pivotal for empowering large language models (LLMs) to generate helpful and harmless responses. However, the performance of preference alignment is highly sensitive to the prevalent noise in the preference data.…

Machine Learning · Computer Science 2024-05-29 Xize Liang , Chao Chen , Shuang Qiu , Jie Wang , Yue Wu , Zhihang Fu , Zhihao Shi , Feng Wu , Jieping Ye

While astonishingly capable, large Language Models (LLM) can sometimes produce outputs that deviate from human expectations. Such deviations necessitate an alignment phase to prevent disseminating untruthful, toxic, or biased information.…

Artificial Intelligence · Computer Science 2024-10-30 Long Tan Le , Han Shu , Tung-Anh Nguyen , Choong Seon Hong , Nguyen H. Tran

In this paper, we introduce \emph{refined Direct Preference Optimization} (rDPO), a method for improving the behavioral alignment of Large Language Models (LLMs) without the need for human-annotated data. The method involves creating…

Computation and Language · Computer Science 2024-02-14 Víctor Gallego

Reinforcement learning (RL) is effective in enhancing the accuracy of large language models in complex reasoning tasks. Existing RL policy optimization frameworks rely on final-answer correctness as feedback signals and rarely capture the…

Artificial Intelligence · Computer Science 2026-04-13 Jinghan Zhang , Fengran Mo , Tharindu Cyril Weerasooriya , Ruimin Dai , Xiaoyan Han , Yanjie Fu , Dakuo Wang , Kunpeng Liu

Reinforcement learning from human feedback (RLHF) has emerged as the standard paradigm for aligning large language models with human preferences. However, reward-based methods grounded in the Bradley-Terry assumption struggle to capture the…

Artificial Intelligence · Computer Science 2026-04-08 Fang Wu , Xu Huang , Weihao Xuan , Zhiwei Zhang , Yijia Xiao , Guancheng Wan , Xiaomin Li , Bing Hu , Peng Xia , Jure Leskovec , Yejin Choi

Direct Preference Optimization (DPO) has gained significant attention for its simplicity and computational efficiency in aligning large language models (LLMs). Recent advancements have extended DPO to multimodal scenarios, achieving strong…

Computation and Language · Computer Science 2025-05-27 Yeyuan Wang , Dehong Gao , Rujiao Long , Lei Yi , Linbo Jin , Libin Yang , Xiaoyan Cai

Group Relative Policy Optimization (GRPO) is highly effective for post-training autoregressive (AR) language models, yet its direct application to diffusion large language models (dLLMs) often triggers reward collapse. We identify two…

Machine Learning · Computer Science 2026-03-10 Jianyuan Zhong , Kaibo Wang , Ding Ding , Zijin Feng , Haoli Bai , Yang Xiang , Jiacheng Sun , Qiang Xu

On-policy reinforcement learning (RL), particularly Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO), has become the dominant paradigm for fine-tuning large language models (LLMs). While policy ratio clipping…

Machine Learning · Computer Science 2026-01-08 Yu Luo , Shuo Han , Yihan Hu , Dong Li , Jianye Hao

Recently, self-play fine-tuning (SPIN) has been proposed to adapt large language models to downstream applications with scarce expert-annotated data, by iteratively generating synthetic responses from the model itself. However, SPIN is…

Computation and Language · Computer Science 2026-01-14 Yibo Wang , Hai-Long Sun , Qing-Guo Chen , Zhao Xu , Weihua Luo , Kaifu Zhang , Lijun Zhang

Improving the alignment of language models with human preferences remains an active research challenge. Previous approaches have primarily utilized Reinforcement Learning from Human Feedback (RLHF) via online RL methods such as Proximal…

Computation and Language · Computer Science 2024-01-25 Tianqi Liu , Yao Zhao , Rishabh Joshi , Misha Khalman , Mohammad Saleh , Peter J. Liu , Jialu Liu

The enhancement of reasoning capabilities in large language models (LLMs) has garnered significant attention, with supervised fine-tuning (SFT) and reinforcement learning emerging as dominant paradigms. While recent studies recognize the…

Artificial Intelligence · Computer Science 2026-03-17 Zhijie Wang

Proximal Policy Optimization (PPO) is commonly used in Reinforcement Learning from Human Feedback to align large language models (LLMs) with downstream tasks. This paper investigates the feasibility of using PPO for direct reinforcement…

Computation and Language · Computer Science 2024-10-23 Alexander G. Padula , Dennis J. N. J. Soemers

Group-Relative Policy Optimization (GRPO) is a key technique for training large reasoning models, yet it suffers from a critical vulnerability: the \emph{Think-Answer Mismatch}, where noisy reward signals corrupt the learning process. This…

Machine Learning · Computer Science 2025-08-11 Si Shen , Peijun Shen , Wenhua Zhao , Danhao Zhu

How can Large Language Models (LLMs) be aligned with human intentions and values? A typical solution is to gather human preference on model outputs and finetune the LLMs accordingly while ensuring that updates do not deviate too far from a…

Computation and Language · Computer Science 2024-05-28 Hung Le , Quan Tran , Dung Nguyen , Kien Do , Saloni Mittal , Kelechi Ogueji , Svetha Venkatesh

Aligning generative models with human preference via RLHF typically suffers from overoptimization, where an imperfectly learned reward model can misguide the generative model to output undesired responses. We investigate this problem in a…

Machine Learning · Computer Science 2024-12-05 Zhihan Liu , Miao Lu , Shenao Zhang , Boyi Liu , Hongyi Guo , Yingxiang Yang , Jose Blanchet , Zhaoran Wang