A Deep Dive Into Large Language Model Code Generation Mistakes: What and Why?
This work addresses code quality issues in LLM-based automated code generation for software developers, but it is incremental as it builds on prior research by extending analysis to real-world contexts.
The paper tackled the problem of non-syntactic mistakes in LLM-generated code by identifying seven categories, four of which were previously overlooked, and found that GPT-4 with ReAct prompting achieved an F1 score of up to 0.65 in detecting reasons for these mistakes.
Recent advancements in Large Language Models (LLMs) have led to their widespread application in automated code generation. However, these models can still generate defective code that deviates from the specification. Previous research has mainly focused on the mistakes in LLM-generated standalone functions, overlooking real-world software development situations where the successful generation of the code requires software contexts such as external dependencies. In this paper, we considered both of these code generation situations and identified a range of \textit{non-syntactic mistakes} arising from LLMs' misunderstandings of coding question specifications. Seven categories of non-syntactic mistakes were identified through extensive manual analyses, four of which were missed by previous works. To better understand these mistakes, we proposed six reasons behind these mistakes from various perspectives. Moreover, we explored the effectiveness of LLMs in detecting mistakes and their reasons. Our evaluation demonstrated that GPT-4 with the ReAct prompting technique can achieve an F1 score of up to 0.65 when identifying reasons for LLM's mistakes, such as misleading function signatures. We believe that these findings offer valuable insights into enhancing the quality of LLM-generated code.