English
Related papers

Related papers: Zeroth-Order Policy Gradient for Reinforcement Lea…

200 papers

Aligning large language models (LLMs) with human values and intentions is crucial for their utility, honesty, and safety. Reinforcement learning from human feedback (RLHF) is a popular approach to achieve this alignment, but it faces…

Machine Learning · Computer Science 2025-07-22 Junkang Wu , Xue Wang , Zhengyi Yang , Jiancan Wu , Jinyang Gao , Bolin Ding , Xiang Wang , Xiangnan He

Reinforcement learning from human feedback (RLHF) has become a cornerstone for aligning large language models with human preferences. However, the heterogeneity of human feedback, driven by diverse individual contexts and preferences, poses…

Machine Learning · Statistics 2026-03-05 Seong Jin Lee , Will Wei Sun , Yufeng Liu

Reinforcement learning (RL) faces challenges in evaluating policy trajectories within intricate game tasks due to the difficulty in designing comprehensive and precise reward functions. This inherent difficulty curtails the broader…

Artificial Intelligence · Computer Science 2024-07-02 Zichao Shen , Tianchen Zhu , Qingyun Sun , Shiqi Gao , Jianxin Li

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…

While Reinforcement Learning from Human Feedback (RLHF) effectively aligns pretrained Large Language and Vision-Language Models (LLMs, and VLMs) with human preferences, its computational cost and complexity hamper its wider adoption. To…

A critical component of the current generation of language models is preference alignment, which aims to precisely control the model's behavior to meet human needs and values. The most notable among such methods is Reinforcement Learning…

Artificial Intelligence · Computer Science 2024-10-22 Oh Joon Kwon , Daiki E. Matsunaga , Kee-Eung Kim

Preference learning algorithms (e.g., RLHF and DPO) are frequently used to steer LLMs to produce generations that are more preferred by humans, but our understanding of their inner workings is still limited. In this work, we study the…

Machine Learning · Computer Science 2024-11-01 Angelica Chen , Sadhika Malladi , Lily H. Zhang , Xinyi Chen , Qiuyi Zhang , Rajesh Ranganath , Kyunghyun Cho

Aligning Large Language Models (LLMs) to human preferences in content, style, and presentation is challenging, in part because preferences are varied, context-dependent, and sometimes inherently ambiguous. While successful, Reinforcement…

Machine Learning · Computer Science 2024-10-29 Sam Houliston , Alizée Pace , Alexander Immer , Gunnar Rätsch

Reinforcement Learning from Human Feedback (RLHF) is a pivotal technique for aligning large language models (LLMs) with human preferences, yet it is susceptible to reward overoptimization, in which policy models overfit to the reward model,…

The success of AI assistants based on Language Models (LLMs) hinges on Reinforcement Learning from Human Feedback (RLHF) to comprehend and align with user intentions. However, traditional alignment algorithms, such as PPO, are hampered by…

Computation and Language · Computer Science 2024-07-03 Songyang Gao , Qiming Ge , Wei Shen , Shihan Dou , Junjie Ye , Xiao Wang , Rui Zheng , Yicheng Zou , Zhi Chen , Hang Yan , Qi Zhang , Dahua Lin

Reinforcement Learning from Human Feedback (RLHF) is a widely adopted approach for aligning large language models with human values. However, RLHF relies on a reward model that is trained with a limited amount of human preference data,…

Machine Learning · Computer Science 2024-10-23 Shun Zhang , Zhenfang Chen , Sunli Chen , Yikang Shen , Zhiqing Sun , Chuang Gan

Direct Preference Optimization (DPO) has been proposed as a promising alternative to Proximal Policy Optimization (PPO) based Reinforcement Learning with Human Feedback (RLHF). However, empirical evaluations consistently reveal suboptimal…

Machine Learning · Computer Science 2025-03-03 Qinwei Ma , Jingzhe Shi , Can Jin , Jenq-Neng Hwang , Serge Belongie , Lei Li

Large language models (LLMs) have formulated a blueprint for the advancement of artificial general intelligence. Its primary objective is to function as a human-centric (helpful, honest, and harmless) assistant. Alignment with humans…

Language model (LM) post-training (or alignment) involves maximizing a reward function that is derived from preference annotations. Direct Preference Optimization (DPO) is a popular offline alignment method that trains a policy directly on…

Machine Learning · Computer Science 2025-03-04 Adam Fisch , Jacob Eisenstein , Vicky Zayats , Alekh Agarwal , Ahmad Beirami , Chirag Nagpal , Pete Shaw , Jonathan Berant

We introduce Adversarial Policy Optimization (AdvPO), a novel solution to the pervasive issue of reward over-optimization in Reinforcement Learning from Human Feedback (RLHF) for Large Language Models (LLMs). Over-optimization occurs when a…

Machine Learning · Computer Science 2024-07-10 Xiaoying Zhang , Jean-Francois Ton , Wei Shen , Hongning Wang , Yang Liu

Collecting high-quality preference datasets for reinforcement learning from human feedback (RLHF) is resource-intensive and challenging. As a result, researchers often train reward models on extensive offline datasets which aggregate…

Machine Learning · Computer Science 2024-12-17 Shambhavi Krishna , Aishwarya Sahoo

Direct Preference Optimization (DPO) simplifies reinforcement learning from human feedback (RLHF) for large language models (LLMs) by directly optimizing human preferences without an explicit reward model. We find that during DPO training,…

Computation and Language · Computer Science 2026-01-01 Junshu Pan , Wei Shen , Shulin Huang , Qiji Zhou , Yue Zhang

Reinforcement Learning algorithms that learn from human feedback (RLHF) need to be efficient in terms of statistical complexity, computational complexity, and query complexity. In this work, we consider the RLHF setting where the feedback…

Machine Learning · Computer Science 2024-03-14 Runzhe Wu , Wen Sun

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 from human feedback (RLHF) has emerged as a powerful technique to make large language models (LLMs) more capable in complex settings. RLHF proceeds as collecting human preference data, training a reward model on said…

Machine Learning · Computer Science 2024-02-05 Nathan Lambert , Roberto Calandra