English
Related papers

Related papers: Smaug: Fixing Failure Modes of Preference Optimisa…

200 papers

DPO (Direct Preference Optimization) has become a widely used offline preference optimization algorithm due to its simplicity and training stability. However, DPO is prone to overfitting and collapse. To address these challenges, we propose…

Machine Learning · Computer Science 2025-08-26 Rui Wang , Qianguo Sun , Chao Song , Junlong Wu , Tianrong Chen , Zhiyun Zeng , Yu Li

Direct Preference Optimization (DPO) has emerged as an effective approach for aligning large language models (LLMs) with human preferences. However, its performance is highly dependent on the quality of the underlying human preference data.…

Machine Learning · Computer Science 2026-03-10 Zixuan Huang , Yikun Ban , Lean Fu , Xiaojie Li , Zhongxiang Dai , Jianxin Li , Deqing Wang

We propose the intermediate direct preference optimization (DPO) method to calculate the DPO loss at selected intermediate layers as an auxiliary loss for finetuning large language models (LLMs). The conventional DPO method fine-tunes a…

Computation and Language · Computer Science 2024-08-07 Atsushi Kojima

A major challenge in aligning large language models (LLMs) with human preferences is the issue of distribution shift. LLM alignment algorithms rely on static preference datasets, assuming that they accurately represent real-world user…

Machine Learning · Computer Science 2026-01-16 Zaiyan Xu , Sushil Vemuri , Kishan Panaganti , Dileep Kalathil , Rahul Jain , Deepak Ramachandran

We introduce ConfPO, a method for preference learning in Large Language Models (LLMs) that identifies and optimizes preference-critical tokens based solely on the training policy's confidence, without requiring any auxiliary models or…

Computation and Language · Computer Science 2025-06-13 Hee Suk Yoon , Eunseop Yoon , Mark Hasegawa-Johnson , Sungwoong Kim , Chang D. Yoo

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

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

Aligning Large Language Models (LLMs) with human feedback is crucial for their development. Existing preference optimization methods such as DPO and KTO, while improved based on Reinforcement Learning from Human Feedback (RLHF), are…

Computation and Language · Computer Science 2024-12-23 Shuo Xie , Fangzhi Zhu , Jiahui Wang , Lulu Wen , Wei Dai , Xiaowei Chen , Junxiong Zhu , Kai Zhou , Bo Zheng

Direct Preference Optimization (DPO), a standard method for aligning language models with human preferences, is traditionally applied to offline preferences. Recent studies show that DPO benefits from iterative training with online…

Computation and Language · Computer Science 2024-06-18 Jie Liu , Zhanhui Zhou , Jiaheng Liu , Xingyuan Bu , Chao Yang , Han-Sen Zhong , Wanli Ouyang

Preference optimization techniques, such as Direct Preference Optimization (DPO), are frequently employed to enhance the reasoning capabilities of large language models (LLMs) in domains like mathematical reasoning and coding, typically…

Computation and Language · Computer Science 2025-02-17 Fangkai Jiao , Geyang Guo , Xingxing Zhang , Nancy F. Chen , Shafiq Joty , Furu Wei

Direct Preference Optimization (DPO) is an effective technique that leverages pairwise preference data (usually one chosen and rejected response pair per user prompt) to align LLMs to human preferences. In practice, multiple responses can…

Computation and Language · Computer Science 2024-11-11 Pulkit Pattnaik , Rishabh Maheshwary , Kelechi Ogueji , Vikas Yadav , Sathwik Tejaswi Madhusudhan

Direct Preference Optimization (DPO) has emerged as a more computationally efficient alternative to Reinforcement Learning from Human Feedback (RLHF) with Proximal Policy Optimization (PPO), eliminating the need for reward models and online…

Computation and Language · Computer Science 2024-10-28 Xin Mao , Feng-Lin Li , Huimin Xu , Wei Zhang , Wang Chen , Anh Tuan Luu

Direct Preference Optimization (DPO) has become a popular approach for aligning language models using pairwise preferences. However, in practical post-training pipelines, on-policy generation typically yields multiple candidate responses…

Machine Learning · Computer Science 2025-06-23 Taneesh Gupta , Rahul Madhavan , Xuchao Zhang , Nagarajan Natarajan , Chetan Bansal , Saravan Rajmohan

As Large Language Models (LLMs) demonstrate remarkable capabilities learned from vast corpora, concerns regarding data privacy and safety are receiving increasing attention. LLM unlearning, which aims to remove the influence of specific…

Machine Learning · Computer Science 2025-10-07 Kai Qin , Jiaqi Wu , Jianxiang He , Haoyuan Sun , Yifei Zhao , Bin Liang , Yongzhe Chang , Tiantian Zhang , Houde Liu

Direct preference optimization (DPO) has emerged as a promising approach for aligning large language models (LLMs) with human preferences. However, the widespread reliance on the response-level Bradley-Terry (BT) model may limit its full…

Artificial Intelligence · Computer Science 2026-02-11 Masanari Oi , Mahiro Ukai , Masahiro Kaneko , Naoaki Okazaki , Nakamasa Inoue

Direct Preference Optimization (DPO) has shown strong potential for mitigating hallucinations in Multimodal Large Language Models (MLLMs). However, existing multimodal DPO approaches often suffer from overfitting due to the difficulty…

Artificial Intelligence · Computer Science 2026-01-05 Longtian Qiu , Shan Ning , Chuyu Zhang , Jiaxuan Sun , Xuming He

Aligning Large Language Models (LLMs) with human preferences is crucial for safe and effective AI interactions. While popular methods like Direct Preference Optimization (DPO) have simplified alignment, they remain sensitive to data noise…

Artificial Intelligence · Computer Science 2026-03-03 Ning Yang , Hai Lin , Yibo Liu , Baoliang Tian , Guoqing Liu , Haijun Zhang

Text-to-image diffusion models deliver high-quality images, yet aligning them with human preferences remains challenging. We revisit diffusion-based Direct Preference Optimization (DPO) for these models and identify a critical pathology:…

Computer Vision and Pattern Recognition · Computer Science 2025-12-03 Minghao Fu , Guo-Hua Wang , Tianyu Cui , Qing-Guo Chen , Zhao Xu , Weihua Luo , Kaifu Zhang

Preference optimization has become a central paradigm for aligning large language models with human feedback. Direct Preference Optimization (DPO) simplifies reinforcement learning from human feedback by directly optimizing pairwise…

Machine Learning · Computer Science 2026-05-05 Inoussa Mouiche

Direct Preference Optimization (DPO) is a powerful paradigm for aligning Large Language Models (LLMs) to human preferences in Machine Translation (MT), but current methods are hindered by two fundamental challenges: (1) flawed reward…

Computation and Language · Computer Science 2025-10-16 Hao Wang , Linlong Xu , Heng Liu , Yangyang Liu , Xiaohu Zhao , Bo Zeng , Liangying Shao , Longyue Wang , Weihua Luo , Kaifu Zhang