Context-DPO: Aligning Language Models for Context-Faithfulness
Baolong Bi, Shaohan Huang, Yiwei Wang, Tianchi Yang, Zihan Zhang, Haizhen Huang, Lingrui Mei, Junfeng Fang
Abstract
Reliable responses from large language models (LLMs) require adherence to user instructions and retrieved information. While alignment techniques help LLMs align with human intentions and values, improving context-faithfulness through alignment remains underexplored. To address this, we propose Context-DPO, the first alignment method specifically designed to enhance LLMs’ context-faithfulness. We introduce ConFiQA, a benchmark that simulates Retrieval-Augmented Generation (RAG) scenarios with knowledge conflicts to evaluate context-faithfulness. By leveraging faithful and stubborn responses to questions with provided context from ConFiQA, our Context-DPO aligns LLMs through direct preference optimization. Extensive experiments demonstrate that our Context-DPO significantly improves context-faithfulness, achieving 35% to 280% improvements on popular open-source models. Further analysis demonstrates that Context-DPO preserves LLMs’ generative capabilities while providing interpretable insights into context utilization.
BibTeX
@inproceedings{bi-etal-2025-context,
title = "Context-{DPO}: Aligning Language Models for Context-Faithfulness",
author = "Bi, Baolong and
Huang, Shaohan and
Wang, Yiwei and
Yang, Tianchi and
Zhang, Zihan and
Huang, Haizhen and
Mei, Lingrui and
Fang, Junfeng and
Li, Zehao and
Wei, Furu and
Deng, Weiwei and
Sun, Feng and
Zhang, Qi and
Liu, Shenghua",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.536/",
doi = "10.18653/v1/2025.findings-acl.536",
pages = "10280--10300",
ISBN = "979-8-89176-256-5"
}