@inproceedings{di-etal-2026-code,
    title = "Code Reffix: A Benchmark for Reflection-Guided Code Repair with Large Language Models",
    author = "Di, Zaiyuan  and
      Chen, Jianting  and
      Yang, Yunxiao  and
      Gao, Xiaoying  and
      Yang, Li  and
      Wang, Zhihao  and
      Xiang, Yang",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.762/",
    pages = "15541--15561",
    ISBN = "979-8-89176-395-1",
    abstract = "While recent studies have increasingly emphasized the role of reflection in code repair tasks, existing benchmarks still target the repair generation capability of LLMs, lacking fine-grained evaluation of reflection generation capability. To this end, we propose Code Reffix, a benchmark featuring an automated pipeline with oracle reflections and a dual-task protocol to decouple the evaluation of reflection from repair. Through extensive experiments on 14 LLMs and fine-tuning analysis, we aim to pinpoint performance bottlenecks of code repair, quantify reflection quality, and verify the value of reflection optimization. Evaluations reveal that underperforming reflection capabilities of small-scale LLMs remain a major bottleneck for code repair. By quantifying this gap, Code Reffix provides a critical foundation for optimizing LLMs to achieve superior repair performance."
}