Analysis on LLMs Performance for Code Summarization

TL;DR

This paper compares the performance of several open-source Large Language Models on code summarization tasks, evaluating their strengths and weaknesses using standard metrics to guide future improvements.

Contribution

It provides a comparative analysis of LLaMA-3, Phi-3, Mistral, and Gemma for code summarization, highlighting their relative effectiveness and limitations.

Findings

LLaMA-3 outperforms others on BLEU and ROUGE metrics.

Different models show varied strengths in summary accuracy.

Insights support better model selection for software engineering tools.

Abstract

Code summarization aims to generate concise natural language descriptions for source code. Deep learning has been used more and more recently in software engineering, particularly for tasks like code creation and summarization. Specifically, it appears that the most current Large Language Models with coding perform well on these tasks. Large Language Models (LLMs) have significantly advanced the field of code summarization, providing sophisticated methods for generating concise and accurate summaries of source code. This study aims to perform a comparative analysis of several open-source LLMs, namely LLaMA-3, Phi-3, Mistral, and Gemma. These models' performance is assessed using important metrics such as BLEU\textsubscript{3.1} and ROUGE\textsubscript{3.2}. Through this analysis, we seek to identify the strengths and weaknesses of each model, offering insights into their applicability and effectiveness in code summarization tasks. Our findings contribute to the ongoing development and refinement of LLMs, supporting their integration into tools that enhance software development and maintenance processes.