@inproceedings{wu-etal-2026-coig,
    title = "{COIG}-{P}: A High-Quality and Large-Scale {C}hinese Preference Dataset for Alignment with Human Values",
    author = "Wu, Siwei  and
      Ren, JinCheng  and
      Du, Xeron  and
      Guo, Shuyue  and
      Qu, Xingwei  and
      Liang, Yiming  and
      Liu, Jie  and
      Li, Yunwen  and
      Loakman, Tyler  and
      Zheng, Tianyu  and
      Feng, Boyu  and
      Yuan, Huaqing  and
      Wang, Zili  and
      Liu, Jiaheng  and
      Huang, Wenhao  and
      Cai, Chenglin  and
      Que, Haoran  and
      Yang, Jian  and
      Bai, Yuelin  and
      Wang, Zekun Moore  and
      Yu, Zhouliang  and
      Lin, Qunshu  and
      Pan, Ding  and
      Jiang, Yuchen Eleanor  and
      Wang, Tiannan  and
      Zhou, Wangchunshu  and
      Wang, Shenzhi  and
      Bu, Xingyuan  and
      Liu, Minghao  and
      Wang, Guoyin  and
      Zhang, Ge  and
      Lin, Chenghua",
    editor = "Demberg, Vera  and
      Inui, Kentaro  and
      Marquez, Llu{\'i}s",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EACL} 2026",
    month = mar,
    year = "2026",
    address = "Rabat, Morocco",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-eacl/2026.findings-eacl.288/",
    pages = "5420--5447",
    ISBN = "979-8-89176-386-9",
    abstract = "Existing Chinese preference datasets suffer from limited scale, restricted domain coverage, and insufficiently rigorous data validation. Human annotation significantly limits the scalability of human preference datasets. As a result, Chinese Alignment and Chinese Reward Models (CRM) have not yet been thoroughly explored. To address these challenges, we design an LLM-based data annotation pipeline with no human intervention. Based on this pipeline, we curate COIG-P (Chinese Open Instruction Generalist - Preference), a high-quality, large-scale Chinese preference dataset consisting of 1M Chinese preference pairs and 92k carefully curated Chinese queries across diverse domains, including Chat, Coding, Maths, and others. We conduct experiments to verify the quality of COIG-P from two perspectives. (1) COIG-P brings significant performance improvements for the Qwen2/2.5 and Infinity-Instruct model series on AlignBench through DPO, with gains ranging from 2{\%} to 12{\%}. Furthermore, it significantly outperforms other existing Chinese preference datasets. (2) We train an 8B-sized CRM and manually annotate a Chinese Reward Benchmark (CRBench). Our CRM demonstrates robust scoring ability on CRBench. In addition, in practical data construction experiments, the quality of the data constructed by our CRM is comparable to that produced by GPT-4o."
}