LLMCup: Ranking-Enhanced Comment Updating with LLMs
Hua Ge, Juan Zhai, Minxue Pan, Fusen He, Ziyue Tan · 2025
Read LLMCup: Ranking-Enhanced Comment Updating with LLMs by Hua Ge, Juan Zhai, Minxue Pan, Fusen He, Ziyue Tan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-08671.txt.
Description
Code comments that lag behind code changes can be updated by language models, Hua Ge and co-authors show in a 2025 paper. Their system is called LLMCup. Comments are essential for readability and maintenance, yet developers update code far more willingly than comments, which leaves outdated documentation that hinders future understanding of a project by new team members.
Prior methods use neural sequence models or heuristic rules, and they can miss or misread crucial information in complex updates. Language models show strong potential to capture the logic behind a code change. The open problem is selecting an appropriate prompt strategy for each update case. LLMCup adds a ranking component to that choice, aiming for accurate updates across varied scenarios.
Catalog record
| CREATOR | Hua Ge, Juan Zhai, Minxue Pan, Fusen He, Ziyue Tan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-08671