Large Language Models (LLMs) have demonstrated unprecedented capabilities in code generation. However, there remains a limited understanding of code generation errors that LLMs can produce. To bridge the gap, we conducted an in-depth analysis of code generation errors across six representative LLMs on the HumanEval dataset. Specifically, we first employed open coding and thematic analysis to distill a comprehensive taxonomy of code generation errors. We analyzed two dimensions of error characteristics -- semantic characteristics and syntactic characteristics. Our analysis revealed that LLMs often made non-trivial, multi-line code generation errors in various locations and with various root causes. We further analyzed the correlation between these errors and task complexity as well as test pass rate. Our findings highlighted several challenges in locating and fixing code generation errors made by LLMs. In the end, we discussed several future directions to address these challenges.
@article{arxiv.2406.08731,
title = {Towards Understanding the Characteristics of Code Generation Errors Made by Large Language Models},
author = {Zhijie Wang and Zijie Zhou and Da Song and Yuheng Huang and Shengmai Chen and Lei Ma and Tianyi Zhang},
journal= {arXiv preprint arXiv:2406.08731},
year = {2025}
}
Comments
To appear in the 47th IEEE/ACM Conference on Software Engineering (ICSE 2025). The first three authors contributed equally to this work