@inproceedings{yu-etal-2025-self-error,
    title = "Self-Error-Instruct: Generalizing from Errors for {LLM}s Mathematical Reasoning",
    author = "Yu, Erxin  and
      Li, Jing  and
      Liao, Ming  and
      Zhu, Qi  and
      Xue, Boyang  and
      Xu, Minghui  and
      Wang, Baojun  and
      Hong, Lanqing  and
      Mi, Fei  and
      Shang, Lifeng",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.417/",
    pages = "8504--8519",
    ISBN = "979-8-89176-251-0",
    abstract = "Although large language models demonstrate strong performance across various domains, they still struggle with numerous bad cases in mathematical reasoning. Previous approaches to learning from errors synthesize training data by solely extrapolating from isolated bad cases, thereby failing to generalize the extensive patterns inherent within these cases. This paper presents Self-Error-Instruct (SEI), a framework that addresses these model weaknesses and synthesizes more generalized targeted training data. Specifically, we explore a target model on two mathematical datasets, GSM8K and MATH, to pinpoint bad cases. Then, we generate error keyphrases for these cases based on the instructor model{'}s (GPT-4o) analysis and identify error types by clustering these keyphrases. Next, we sample a few bad cases during each generation for each identified error type and input them into the instructor model, which synthesizes additional training data using a self-instruct approach. This new data is refined through a one-shot learning process to ensure that only the most effective examples are kept. Finally, we use these curated data to fine-tune the target model, iteratively repeating the process to enhance performance. We apply our framework to various models and observe improvements in their reasoning abilities across both in-domain and out-of-domain mathematics datasets. These results demonstrate the effectiveness of self-error instruction in improving LLMs' mathematical reasoning through error generalization."
}