license: apache-2.0
task_categories:
- text-generation
language:
- en
tags:
- code
size_categories:
- 10K<n<100K
Introduction
We construct the first dataset, named ACPR (Accuracy-Consistency Program Repair) for AdaPR (Adatpvie Program Repair) task, which aims to evaluate the generated patches from accuracy (i.e., fixing bugs correctly) and consistency (i.e., minimizing modifications). For a given buggy program, we pair it with a randomly selected failed test case (a test case includes a test input and an expected output) as well as a passed program from the same user’s submission for the same programming problem, making a ⟨buggy code, f ailed test case, passed code⟩ triplet sample. The whole dataset contains 52,168 triplet data samples. We then split our dataset into train/validation/test sets by the ratio of 8:1:1, ensuring that any particular programming problem appears in only one of them to avoid data leakage problems. To prevent overfitting code data from the same programming problem and to ensure fairness in evaluation, we balance the dataset by capping the maximum number of pairs per problem at 150/10/20 in the train/validation/test sets.
code link of code execution and evaluation
https://github.com/zhenlongDai/AdaPatcher
paper link
https://arxiv.org/pdf/2503.06510
Citation
@article{dai2025less,
title={Less is More: Adaptive Program Repair with Bug Localization and Preference Learning},
author={Dai, Zhenlong and Chen, Bingrui and Zhao, Zhuoluo and Tang, Xiu and Wu, Sai and Yao, Chang and Gao, Zhipeng and Chen, Jingyuan},
journal={arXiv preprint arXiv:2503.06510},
year={2025}
}