A RAG Method for Source Code Inquiry Tailored to Long-Context LLMs
Fuente:
arXiv
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911832782929920 |
|---|---|
| author | Kamiya, Toshihiro |
| author_facet | Kamiya, Toshihiro |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_06082 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A RAG Method for Source Code Inquiry Tailored to Long-Context LLMs Kamiya, Toshihiro Software Engineering 68-04 D.2.3; D.2.5 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. |
| title | A RAG Method for Source Code Inquiry Tailored to Long-Context LLMs |
| topic | Software Engineering 68-04 D.2.3; D.2.5 |
| url | https://arxiv.org/abs/2404.06082 |