Sep 2026
Analyzing and fixing code generation errors of foundation large language models
The goal is to understand the code generation errors of foundation LLMs and explore the solution to resolve directly fixable errors, and to design and evaluate the LlmFix fixing method and constructed the LlmErrorEval dataset.
Hao Wen, Yue-Heng Zhu, Chao Liu et al.
· Empirical Software Engineeri... · 0 citations