A RAG Method for Source Code Inquiry Tailored to Long-Context LLMs

Autor:	Kamiya, Toshihiro
Rok vydání:	2024
Předmět:	Computer Science - Software Engineering D.2.3 D.2.5
Druh dokumentu:	Working Paper
Popis:	Although the context length limitation of large language models (LLMs) has been mitigated, it still hinders their application to software development tasks. This study proposes a method incorporating execution traces into RAG for inquiries about source code. Small-scale experiments confirm a tendency for the method to contribute to improving LLM response quality. Comment: 6 pages, 2 columns, English translation of the manuscript originally presented in Japanese at a domestic workshop
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2404.06082 Zobrazit plný text záznamu View this record from Arxiv