English
Related papers

Related papers: Generalisation of RLHF under Reward Shift and Clip…

200 papers

Reward models trained through Reinforcement Learning from AI Feedback (RLAIF) methods frequently suffer from limited generalizability, which hinders the alignment performance of policy models. This challenge stems from various issues,…

Artificial Intelligence · Computer Science 2026-04-21 Jiaye Lin , Mengdi Li , Xufeng Zhao , Wenhao Lu , Peilin Zhao , Stefan Wermter , Di Wang

A centerpiece of the ever-popular reinforcement learning from human feedback (RLHF) approach to fine-tuning autoregressive language models is the explicit training of a reward model to emulate human feedback, distinct from the language…

Computation and Language · Computer Science 2023-05-22 Wanqiao Xu , Shi Dong , Dilip Arumugam , Benjamin Van Roy

Reinforcement Learning from Human Feedback (RLHF) has become a dominating strategy in aligning Language Models (LMs) with human values/goals. The key to the strategy is learning a reward model ($\varphi$), which can reflect the latent…

Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models (LLMs) with human preferences, thereby enhancing the quality of responses generated. A critical component of RLHF is the reward model,…

Artificial Intelligence · Computer Science 2024-06-25 Yulan Hu , Qingyang Li , Sheng Ouyang , Ge Chen , Kaihui Chen , Lijun Mei , Xucheng Ye , Fuzheng Zhang , Yong Liu

Large language models (LLMs) trained via pretraining and supervised fine-tuning (SFT) can still produce harmful and misaligned outputs, or struggle in domains like math and coding. Reinforcement learning (RL)-based post-training methods,…

Computation and Language · Computer Science 2026-05-19 Zhichao Wang , Kiran Ramnath , Bin Bi , Shiva Kumar Pentyala , Sougata Chaudhuri , Shubham Mehrotra , Zixu , Zhu , Xiang-Bo Mao , Sitaram Asur , Na , Cheng

This paper addresses the challenge of aligning large language models (LLMs) with diverse human preferences within federated learning (FL) environments, where standard methods often fail to adequately represent diverse viewpoints. We…

Computation and Language · Computer Science 2025-12-17 Mahmoud Srewa , Tianyu Zhao , Salma Elmalaki

Mathematical reasoning is a central challenge for large language models (LLMs), requiring not only correct answers but also faithful reasoning processes. Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a promising…

Machine Learning · Computer Science 2025-12-02 Md Tanvirul Alam , Nidhi Rastogi

This work proposes `PET', a novel pessimistic reward fine-tuning method, to learn a pessimistic reward model robust against reward hacking in offline reinforcement learning from human feedback (RLHF). Traditional reward modeling techniques…

Machine Learning · Computer Science 2025-05-28 Yinglun Xu , Hangoo Kang , Tarun Suresh , Yuxuan Wan , Gagandeep Singh

Aligning human preference and value is an important requirement for contemporary foundation models. State-of-the-art techniques such as Reinforcement Learning from Human Feedback (RLHF) often consist of two stages: 1) supervised fine-tuning…

Artificial Intelligence · Computer Science 2024-10-29 Jiaxiang Li , Siliang Zeng , Hoi-To Wai , Chenliang Li , Alfredo Garcia , Mingyi Hong

Current Large Language Models (LLMs) often undergo supervised fine-tuning (SFT) to acquire tool use capabilities. However, SFT struggles to generalize to unfamiliar or complex tool use scenarios. Recent advancements in reinforcement…

Machine Learning · Computer Science 2025-04-22 Cheng Qian , Emre Can Acikgoz , Qi He , Hongru Wang , Xiusi Chen , Dilek Hakkani-Tür , Gokhan Tur , Heng Ji

Reinforcement Learning with Human Feedback (RLHF) is a widely used fine-tuning approach that aligns machine learning model, particularly Language Model (LM) with human preferences. There are typically multiple objectives driving the…

Machine Learning · Computer Science 2025-02-25 Nuoya Xiong , Aarti Singh

Reinforcement Learning from Human Feedback (RLHF) has achieved impressive empirical successes while relying on a small amount of human feedback. However, there is limited theoretical justification for this phenomenon. Additionally, most…

Machine Learning · Computer Science 2024-07-16 Yihan Du , Anna Winnicki , Gal Dalal , Shie Mannor , R. Srikant

Reinforcement Learning from Human Feedback (RLHF) has achieved considerable success in aligning large language models (LLMs) by modeling human preferences with a learnable reward model and employing a reinforcement learning algorithm to…

Machine Learning · Computer Science 2025-05-20 Jianfeng Cai , Jinhua Zhu , Ruopei Sun , Yue Wang , Li Li , Wengang Zhou , Houqiang Li

Reward models have been increasingly critical for improving the reasoning capability of LLMs. Existing research has shown that a well-trained reward model can substantially improve model performances at inference time via search. However,…

Machine Learning · Computer Science 2024-11-28 Jiaxuan Gao , Shusheng Xu , Wenjie Ye , Weilin Liu , Chuyi He , Wei Fu , Zhiyu Mei , Guangju Wang , Yi Wu

Spurious correlations in a reward model's training data can prevent Reinforcement Learning from Human Feedback (RLHF) from identifying the desired goal and induce unwanted behaviors. This paper shows that offline RLHF is susceptible to…

Machine Learning · Computer Science 2024-10-16 Xin Chen , Sam Toyer , Florian Shkurti

Reinforcement Learning with Human Feedback (RLHF) has emerged as a key paradigm for task-specific fine-tuning of language models using human preference data. While numerous publicly available preference datasets provide pairwise comparisons…

Computation and Language · Computer Science 2025-11-03 Ashwin Kumar , Yuzi He , Aram H. Markosyan , Bobbie Chern , Imanol Arrieta-Ibarra

Training multi-modal large language models (MLLMs) that align with human intentions is a long-term challenge. Traditional score-only reward models for alignment suffer from low accuracy, weak generalization, and poor interpretability,…

Artificial Intelligence · Computer Science 2025-05-27 Jiayi Zhou , Jiaming Ji , Boyuan Chen , Jiapeng Sun , Wenqi Chen , Donghai Hong , Sirui Han , Yike Guo , Yaodong Yang

The safety of large language models (LLMs) relies on alignment techniques such as reinforcement learning from human feedback (RLHF). However, recent theoretical analyses suggest that reinforcement learning-based training does not acquire…

Machine Learning · Computer Science 2026-04-06 Haruhi Shida , Koo Imai , Keigo Kansa

Post-training has become a crucial step for unlocking the capabilities of large language models, with reinforcement learning (RL) emerging as a critical paradigm. Recent RL-based post-training has increasingly split into two paradigms:…

Machine Learning · Computer Science 2026-05-18 Shangjian Yin , Yu Fu , Yue Dong , Zhouxing Shi

Large language models are typically aligned with human preferences by optimizing $\textit{reward models}$ (RMs) fitted to human feedback. However, human preferences are multi-faceted, and it is increasingly common to derive reward from a…