English
Related papers

Related papers: What Does Preference Learning Recover from Pairwis…

200 papers

Preference-based reinforcement learning (PbRL) promises to learn a complex reward function with binary human preference. However, such human-in-the-loop formulation requires considerable human effort to assign preference labels to segment…

Machine Learning · Computer Science 2023-07-20 Yachen Kang , Li He , Jinxin Liu , Zifeng Zhuang , Donglin Wang

The standard way to evaluate language models on subjective tasks is through pairwise comparisons: an annotator chooses the "better" of two responses to a prompt. Leaderboards aggregate these comparisons into a single Bradley-Terry (BT)…

Machine Learning · Computer Science 2026-02-26 Hadi Khalaf , Serena L. Wang , Daniel Halpern , Itai Shapira , Flavio du Pin Calmon , Ariel D. Procaccia

We study preferences estimated from finite choice experiments and provide sufficient conditions for convergence to a unique underlying "true" preference. Our conditions are weak, and therefore valid in a wide range of economic environments.…

Theoretical Economics · Economics 2020-11-03 Christopher P. Chambers , Federico Echenique , Nicolas Lambert

Generative Reward Models (GRMs) have demonstrated strong performance in reward modeling, due to their interpretability and potential for refinement through reinforcement learning (RL). However, widely used pairwise GRMs create a…

Reward models trained on human preference data have demonstrated strong effectiveness in aligning Large Language Models (LLMs) with human intent under the framework of Reinforcement Learning from Human Feedback (RLHF). However, RLHF remains…

While preference modelling is becoming one of the pillars of machine learning, the problem of preference explanation remains challenging and underexplored. In this paper, we propose \textsc{Pref-SHAP}, a Shapley value-based model…

Machine Learning · Statistics 2022-11-09 Robert Hu , Siu Lun Chau , Jaime Ferrando Huertas , Dino Sejdinovic

Language models serve as proxies for human preference judgements in alignment and evaluation, yet they exhibit systematic miscalibration, prioritizing superficial patterns over substantive qualities. This bias manifests as overreliance on…

Computation and Language · Computer Science 2026-03-05 Anirudh Bharadwaj , Chaitanya Malaviya , Nitish Joshi , Mark Yatskar

In consumer theory, ranking available objects by means of preference relations yields the most common description of individual choices. However, preference-based models assume that individuals: (1) give their preferences only between pairs…

Machine Learning · Computer Science 2023-02-02 Alessio Benavoli , Dario Azzimonti , Dario Piga

Crowdsourced pairwise evaluation has emerged as a scalable approach for assessing foundation models. However, applying it to Text to Speech(TTS) introduces high variance due to linguistic diversity and multidimensional nature of speech…

A common problem faced in statistical inference is drawing conclusions from paired comparisons, in which two objects compete and one is declared the victor. A probabilistic approach to such a problem is the Bradley-Terry model, first…

Applications · Statistics 2017-12-19 Gabriel C. Phelan , John T. Whelan

Standard methods in preference learning involve estimating the parameters of discrete choice models from data of selections (choices) made by individuals from a discrete set of alternatives (the choice set). While there are many models for…

Machine Learning · Computer Science 2021-08-18 Kiran Tomlinson , Johan Ugander , Austin R. Benson

Reinforcement learning from human feedback (RLHF) has demonstrated remarkable effectiveness in aligning large language models (LLMs) with human preferences. Many existing alignment approaches rely on the Bradley-Terry (BT) model assumption,…

Machine Learning · Computer Science 2025-02-25 Yuheng Zhang , Dian Yu , Tao Ge , Linfeng Song , Zhichen Zeng , Haitao Mi , Nan Jiang , Dong Yu

Direct Preference Optimization (DPO) is a widely used RL-free method for aligning language models from pairwise preferences, but it models preferences over full sequences even though generation is driven by per-token decisions. Existing…

Computation and Language · Computer Science 2026-05-15 Truong Nguyen , Tien-Phat Nguyen , Linh Ngo Van , Duy Minh Ho Nguyen , Khoa D. Doan , Trung Le

Preference-based reinforcement learning (PbRL) bypasses explicit reward engineering by inferring reward functions from human preference comparisons, enabling better alignment with human intentions. However, humans often struggle to label a…

Machine Learning · Computer Science 2025-06-11 Ni Mu , Hao Hu , Xiao Hu , Yiqin Yang , Bo Xu , Qing-Shan Jia

We demonstrate that a wide array of machine learning algorithms are specific instances of one single paradigm: reciprocal learning. These instances range from active learning over multi-armed bandits to self-training. We show that all these…

Machine Learning · Statistics 2024-11-05 Julian Rodemann , Christoph Jansen , Georg Schollmeyer

Preference based Reinforcement Learning (PbRL) removes the need to hand specify a reward function by learning a reward from preference feedback over policy behaviors. Current approaches to PbRL do not address the credit assignment problem…

Machine Learning · Computer Science 2024-04-16 Mudit Verma , Katherine Metcalf

This paper is dedicated to a cautious learning methodology for predicting preferences between alternatives characterized by binary attributes (formally, each alternative is seen as a subset of attributes). By "cautious", we mean that the…

Artificial Intelligence · Computer Science 2022-06-16 Hugo Gilbert , Mohamed Ouaguenouni , Meltem Ozturk , Olivier Spanjaard

We study stochastic optimization with data-adaptive sampling schemes to train pairwise learning models. Pairwise learning is ubiquitous, and it covers several popular learning tasks such as ranking, metric learning and AUC maximization. A…

Machine Learning · Computer Science 2025-04-10 Sijia Zhou , Yunwen Lei , Ata Kabán

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

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