English
Related papers

Related papers: Ambiguity Awareness Optimization: Towards Semantic…

200 papers

We study how data of higher quality can be leveraged to improve performance in Direct Preference Optimization (DPO), aiming to understand its impact on DPO training dynamics. Our analyses show that both the solution space and the…

Machine Learning · Computer Science 2025-10-14 Kyung Rok Kim , Yumo Bai , Chonghuan Wang , Guanting Chen

Direct Preference Optimization (DPO) has been widely used for aligning language models with human preferences in a supervised manner. However, several key questions remain unresolved: the rationale behind its log-ratio reward, how the…

Machine Learning · Computer Science 2025-10-03 Yunjae Won , Hyunji Lee , Hyeonbin Hwang , Minjoon Seo

Aligning large language models (LLMs) with human preferences in federated learning (FL) is challenging due to decentralized, privacy-sensitive, and highly non-IID preference data. Direct Preference Optimization (DPO) offers an efficient…

Machine Learning · Computer Science 2026-03-23 Kewen Zhu , Liping Yi , Zhiming Zhao , Zhuang Qi , Han Yu , Qinghua Hu

Offline preference optimization offers a simpler and more stable alternative to RLHF for aligning language models. However, their effectiveness is critically dependent on ranking accuracy, a metric where further gains are highly impactful.…

Computation and Language · Computer Science 2025-11-18 Ruibo Deng , Duanyu Feng , Wenqiang Lei

Preference optimization methods such as DPO align large language models (LLMs) using paired comparisons, but their effectiveness can be highly sensitive to the quality and difficulty of preference pairs. A common heuristic treats…

Artificial Intelligence · Computer Science 2026-02-03 Jinlong Pang , Zhaowei Zhu , Na Di , Yichi Zhang , Yaxuan Wang , Chen Qian , Yang Liu

DPO has become a widely adopted alternative to RLHF for aligning LLMs with human preferences, eliminating the need for a separate reward model or RL loop. Recent theoretical analysis uncovers an asymmetric gradient behavior in DPO: the loss…

Computation and Language · Computer Science 2026-05-28 Shaolong Chen , Madalina Ciobanu , Qingqing Mao , Ritankar Das

Direct Preference Optimization (DPO) and its variants have become increasingly popular for aligning language models with human preferences. These methods aim to teach models to better distinguish between chosen (or preferred) and rejected…

Computation and Language · Computer Science 2025-06-09 Xiliang Yang , Feng Jiang , Qianen Zhang , Lei Zhao , Xiao Li

Reinforcement learning from human feedback (RLHF) has emerged as a reliable approach to aligning large language models (LLMs) to human preferences. Among the plethora of RLHF techniques, proximal policy optimization (PPO) is of the most…

Computation and Language · Computer Science 2023-11-06 Banghua Zhu , Hiteshi Sharma , Felipe Vieira Frujeri , Shi Dong , Chenguang Zhu , Michael I. Jordan , Jiantao Jiao

Direct Preference Optimization (DPO) has become a prominent method for aligning Large Language Models (LLMs) with human preferences. While DPO has enabled significant progress in aligning English LLMs, multilingual preference alignment is…

Computation and Language · Computer Science 2025-06-06 Wen Yang , Junhong Wu , Chen Wang , Chengqing Zong , Jiajun Zhang

Direct Preference Optimization (DPO) has shown effectiveness in aligning multi-modal large language models (MLLM) with human preferences. However, existing methods exhibit an imbalanced responsiveness to the data of varying hardness,…

Computer Vision and Pattern Recognition · Computer Science 2025-02-12 Jinda Lu , Junkang Wu , Jinghan Li , Xiaojun Jia , Shuo Wang , YiFan Zhang , Junfeng Fang , Xiang Wang , Xiangnan He

Direct Preference Optimization (DPO) improves the alignment of large language models (LLMs) with human values by training directly on human preference datasets, eliminating the need for reward models. However, due to the presence of…

Artificial Intelligence · Computer Science 2024-06-11 Biqing Qi , Pengfei Li , Fangyuan Li , Junqi Gao , Kaiyan Zhang , Bowen Zhou

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

Recent advancements in human preference optimization, originally developed for Large Language Models (LLMs), have shown significant potential in improving text-to-image diffusion models. These methods aim to learn the distribution of…

Computer Vision and Pattern Recognition · Computer Science 2025-10-21 Amir Saeidi , Yiran Luo , Agneet Chatterjee , Shamanthak Hegde , Bimsara Pathiraja , Yezhou Yang , Chitta Baral

Recent advancements in reinforcement learning from human feedback have shown that utilizing fine-grained token-level reward models can substantially enhance the performance of Proximal Policy Optimization (PPO) in aligning large language…

Machine Learning · Computer Science 2025-06-18 Mingkang Zhu , Xi Chen , Zhongdao Wang , Bei Yu , Hengshuang Zhao , Jiaya Jia

Direct Preference Optimization is an offline post-SFT method for aligning language models from preference pairs, with strong results in instruction following and summarization. However, DPO's sequence-level implicit reward can be brittle…

Computation and Language · Computer Science 2026-03-03 Samah Fodeh , Linhai Ma , Ganesh Puthiaraju , Srivani Talakokkul , Afshan Khan , Ashley Hagaman , Sarah R. Lowe , Aimee Kendall Roundtree

With the rapid development of Large Language Models (LLMs), numerous Reinforcement Learning from Human Feedback (RLHF) algorithms have been introduced to improve model safety and alignment with human preferences. These algorithms can be…

Machine Learning · Computer Science 2025-02-06 Xuerui Su , Yue Wang , Jinhua Zhu , Mingyang Yi , Feng Xu , Zhiming Ma , Yuting Liu

The prevalent deployment of learning from human preferences through reinforcement learning (RLHF) relies on two important approximations: the first assumes that pairwise preferences can be substituted with pointwise rewards. The second…

Artificial Intelligence · Computer Science 2023-11-23 Mohammad Gheshlaghi Azar , Mark Rowland , Bilal Piot , Daniel Guo , Daniele Calandriello , Michal Valko , Rémi Munos

We present Anchored Direct Preference Optimization (ADPO), a policy alignment method derived from first principles of KL-regularized reinforcement learning. Unlike standard approaches that treat the reference policy merely as a regularizer,…

Machine Learning · Computer Science 2026-01-13 Wang Zixian

Reinforcement Learning with Human Feedback (RLHF) is the key to the success of large language models (LLMs) in recent years. In this work, we first introduce the concepts of knowledge breadth and knowledge depth, which measure the…

Computation and Language · Computer Science 2025-02-21 Sizhe Wang , Yongqi Tong , Hengyuan Zhang , Dawei Li , Xin Zhang , Tianlong Chen

The widespread application of large language models (LLMs) raises increasing demands on ensuring safety or imposing constraints, such as reducing harmful content and adhering to predefined rules. While there have been several works studying…

Machine Learning · Computer Science 2026-02-13 Yihan Du , Seo Taek Kong , R. Srikant