BOAD: Hierarchical SWE Agents via Bandit Optimization
Published:Dec 29, 2025 17:41
•1 min read
•ArXiv
Analysis
This paper addresses the limitations of single-agent LLM systems in complex software engineering tasks by proposing a hierarchical multi-agent approach. The core contribution is the Bandit Optimization for Agent Design (BOAD) framework, which efficiently discovers effective hierarchies of specialized sub-agents. The results demonstrate significant improvements in generalization, particularly on out-of-distribution tasks, surpassing larger models. This work is important because it offers a novel and automated method for designing more robust and adaptable LLM-based systems for real-world software engineering.
Key Takeaways
- •Proposes BOAD, a novel framework for automatically discovering hierarchical multi-agent systems for software engineering.
- •Formulates hierarchy discovery as a multi-armed bandit problem for efficient exploration.
- •Demonstrates superior performance compared to single-agent and manually designed multi-agent systems, especially on out-of-distribution tasks.
- •Achieves state-of-the-art results on SWE-bench-Live, surpassing larger models.
Reference
“BOAD outperforms single-agent and manually designed multi-agent systems. On SWE-bench-Live, featuring more recent and out-of-distribution issues, our 36B system ranks second on the leaderboard at the time of evaluation, surpassing larger models such as GPT-4 and Claude.”