Get started
Work

PLOT: Enhancing Preference Learning via Optimal Transport

Year: 2026
Type: preprint
Abstract: Preference learning in Large Language Models (LLMs) has advanced significantly, yet existing methods remain limited by modest performance gains, high computational costs, hyperparameter sensitivity, and insufficient modeling of global token-level relationships. We introduce PLOT, which enhances Preference Learning in fine-tuning-based alignment through a token-level loss derived from Optimal Transport. By formulating preference learning as an Optimal Transport Problem, PLOT aligns model outputs with human preferences while preserving the original distribution of LLMs, ensuring stability and robustness. Furthermore, PLOT leverages token embeddings to capture semantic relationships, enabling globally informed optimization. Experiments across two preference categories - Human Values and Logic & Problem Solving - spanning seven subpreferences demonstrate that PLOT consistently improves alignment performance while maintaining fluency and coherence. These results substantiate optimal transport as a principled methodology for preference learning, establishing a theoretically grounded framework that provides new insights for preference learning of LLMs.
DOI: 10.48550/arxiv.2604.01837
Authors: Zhu, Liang, Bai, Yuelin, Ren, Xiankun, Yang, Jiaxi, Zhang, Lei, Fang, Feiteng, Alinejad-Rokny, Hamid, Tan, Minghuan, Yang, Min
Institutions: Chinese Academy of Sciences, Southern University of Science and Technology, Shenzhen Institutes of Advanced Technology, Chongqing University, University of Science and Technology of China, UNSW Sydney
Open Access status: green
OpenAlex ID: W7148846287
Cites:
Cited by:
Related to: