Summary of Coder: Issue Resolving with Multi-agent and Task Graphs, by Dong Chen and Shaoxin Lin and Muhan Zeng and Daoguang Zan and Jian-gang Wang and Anton Cheshkov and Jun Sun and Hao Yu and Guoliang Dong and Artem Aliev and Jie Wang and Xiao Cheng and Guangtai Liang and Yuchi Ma and Pan Bian and Tao Xie and Qianxiang Wang
CodeR: Issue Resolving with Multi-Agent and Task Graphs
by Dong Chen, Shaoxin Lin, Muhan Zeng, Daoguang Zan, Jian-Gang Wang, Anton Cheshkov, Jun Sun, Hao Yu, Guoliang Dong, Artem Aliev, Jie Wang, Xiao Cheng, Guangtai Liang, Yuchi Ma, Pan Bian, Tao Xie, Qianxiang Wang
First submitted to arxiv on: 3 Jun 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Medium Difficulty Summary: GitHub’s recent attention on issue resolving has sparked interest in academia and industry. To address this challenge, SWE-bench is proposed as a metric for evaluating performance. This paper presents CodeR, a multi-agent framework that utilizes pre-defined task graphs to repair bugs and add features within code repositories. CodeR achieves 28.33% issue resolution on the SWE-bench lite dataset by submitting only once per issue. We investigate the impact of each design aspect on CodeR’s performance and provide insights for advancing this research direction in software engineering, particularly in areas like GitHub issue resolving. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Low Difficulty Summary: A big problem is happening on websites like GitHub where people report bugs or ask for new features. To solve this, we created a system called CodeR that uses teams of “experts” to fix the issues and add new features. We tested CodeR and it was able to fix 28% of these problems just by looking at them once! We looked at how different parts of CodeR worked together and what we can do next to make it even better. |
Keywords
» Artificial intelligence » Attention
