RepoMAS: Solving Progressively Specified Tasks with Issue-Driven Multi-Agent Systems

DOI10.48550/ARXIV.2609.32490MaRDI QIDQ7362017arXivFDO

Authors Muyun Yang, Tiejun Zhao, Yuchen Song, Andong Chen, Wenxin Zhu

Publication date 26 September 2026

Published in arXiv

Tldr We propose RepoMAS, an issue-driven multi-agent framework inspired by open-source project management that records newly discovered task requirements and uses them to revise task specification and execution structure during problem solving. Across ProgSpec and five existing benchmarks, RepoMAS achieves the best performance.

Full work available at URL https://doi.org/10.48550/ARXIV.2609.32490

arXiv classification

cs.AI

This page was built for publication: RepoMAS: Solving Progressively Specified Tasks with Issue-Driven Multi-Agent Systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7362017)