English
Related papers

Related papers: Unintentional Unalignment: Likelihood Displacement…

200 papers

In practice, preference learning from human feedback depends on incomplete data with hidden context. Hidden context refers to data that affects the feedback received, but which is not represented in the data used to train a preference…

Machine Learning · Computer Science 2024-04-18 Anand Siththaranjan , Cassidy Laidlaw , Dylan Hadfield-Menell

A common technique for aligning large language models (LLMs) relies on acquiring human preferences by comparing multiple generations conditioned on a fixed context. This method, however, relies solely on pairwise comparisons, where the…

Computation and Language · Computer Science 2025-01-09 Hritik Bansal , Ashima Suvarna , Gantavya Bhatt , Nanyun Peng , Kai-Wei Chang , Aditya Grover

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 the domain of complex reasoning tasks, such as mathematical reasoning, recent advancements have proposed the use of Direct Preference Optimization (DPO) to suppress output of dispreferred responses, thereby enhancing the long-chain…

Computation and Language · Computer Science 2025-10-27 Weibin Liao , Xu Chu , Yasha Wang

Offline preference optimization methods are efficient for large language models (LLMs) alignment. Direct Preference optimization (DPO)-like learning, one of the most popular approaches, stands out for its efficiency in reward modeling.…

Machine Learning · Computer Science 2026-05-26 Xiaobo Wang , Zixia Jia , Jiaqi Li , Qi Liu , Zilong Zheng

This study addresses the challenge of noise in training datasets for Direct Preference Optimization (DPO), a method for aligning Large Language Models (LLMs) with human preferences. We categorize noise into pointwise noise, which includes…

Machine Learning · Computer Science 2025-04-21 Junkang Wu , Yuexiang Xie , Zhengyi Yang , Jiancan Wu , Jiawei Chen , Jinyang Gao , Bolin Ding , Xiang Wang , Xiangnan He

Audio large language models (Audio LLMs) exhibit systematic failures in transcribing code-switching speech despite strong multilingual capabilities. Focusing on English-Mandarin, we identify three failure modes: language omission,…

Computation and Language · Computer Science 2026-05-26 Trung Nguyen Quang , Cheng Yi Lewis Won , Minh Duc Pham , Yingxu He , Shuo Sun , Ai Ti Aw

Traditional RLHF-based LLM alignment methods explicitly maximize the expected rewards from a separate reward model. More recent supervised alignment methods like Direct Preference Optimization (DPO) circumvent this phase to avoid problems…

Machine Learning · Computer Science 2025-02-03 Abhijnan Nath , Changsoo Jung , Ethan Seefried , Nikhil Krishnaswamy

Large Language Models (LLMs) have demonstrated remarkable potential in automating software development tasks. While recent advances leverage Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) to align models with human…

Software Engineering · Computer Science 2025-12-09 Xin Yin , Chao Ni , Xiaohu Yang

Preference-based feedback is important for many applications in machine learning where evaluation of a reward function is not feasible. Notable recent examples arise in preference alignment for large language models, including in…

As AI systems become more capable, deceptive behaviors can undermine evaluation and mislead users at deployment. Recent work has shown that lie detectors can accurately classify deceptive behavior, but they are not typically used in the…

Machine Learning · Computer Science 2025-11-19 Chris Cundy , Adam Gleave

After pre-training, large language models are aligned with human preferences based on pairwise comparisons. State-of-the-art alignment methods (such as PPO-based RLHF and DPO) are built on the assumption of aligning with a single preference…

Machine Learning · Computer Science 2025-05-30 Paul Gölz , Nika Haghtalab , Kunhe Yang

Large language models exhibit impressive reasoning capabilities yet frequently generate plausible but incorrect solutions, a phenomenon commonly termed hallucination. This paper investigates the effect of training objective composition on…

Machine Learning · Computer Science 2026-01-13 Murtaza Nikzad , Raghuram Ramanujan

The recent success in using human preferences to align large language models (LLMs) has significantly improved their performance in various downstream tasks, such as question answering, mathematical reasoning, and code generation. However,…

Machine Learning · Computer Science 2026-05-18 Xiaoqiang Lin , Arun Verma , Zhongxiang Dai , Daniela Rus , See-Kiong Ng , Bryan Kian Hsiang Low

Reinforcement learning (RL) has become a predominant technique to align language models (LMs) with human preferences or promote outputs which are deemed to be desirable by a given reward function. Standard RL approaches optimize average…

Machine Learning · Computer Science 2025-10-27 Stephen Zhao , Aidan Li , Rob Brekelmans , Roger Grosse

Diffusion language models (dLLMs) are an emerging alternative to autoregressive (AR) generators, but aligning them to human preferences is challenging because sequence log-likelihoods are intractable and pairwise preference data are costly…

Machine Learning · Computer Science 2025-11-13 Vaibhav Jindal , Hejian Sang , Chun-Mao Lai , Yanning Chen , Zhipeng Wang

Neural metrics for machine translation (MT) evaluation have become increasingly prominent due to their superior correlation with human judgments compared to traditional lexical metrics. Researchers have therefore utilized neural metrics…

Computation and Language · Computer Science 2025-11-21 Hippolyte Gisserot-Boukhlef , Ricardo Rei , Emmanuel Malherbe , Céline Hudelot , Pierre Colombo , Nuno M. Guerreiro

Self-generated counterfactual explanations (SCEs) are minimally modified inputs (minimality) generated by large language models (LLMs) that flip their own predictions (validity), offering a causally grounded approach to unraveling black-box…

Computation and Language · Computer Science 2026-05-13 Yilong Wang , Qianli Wang , Bohao Chu , Yihong Liu , Jing Yang , Simon Ostermann

Direct Preference Optimization (DPO) and related methods align large language models from pairwise preferences by regularizing updates against a fixed reference policy. As the policy drifts, a static reference, however, can become…

Machine Learning · Computer Science 2026-05-18 Youngjae Cho , Jongsuk Kim , Ji-Hoon Kim

Direct preference optimization (DPO) has shown to be an effective method for large language model (LLM) alignment. Recent works have attempted to apply DPO to multimodal scenarios but have found it challenging to achieve consistent…

Computer Vision and Pattern Recognition · Computer Science 2024-10-08 Fei Wang , Wenxuan Zhou , James Y. Huang , Nan Xu , Sheng Zhang , Hoifung Poon , Muhao Chen