ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement
Researchers propose ModularRSI, a modular benchmark-disjoint recursive self-improvement framework that evolves agent harnesses across five modules, improving TB2.0 and SWE-Bench Verified results.
ModularRSI targets generalizable recursive self-improvement (RSI) for agent harnesses by contrasting successful and failed trajectories for the same task and aggregating evidence across tasks to find recurring behavioral deficiencies. It decomposes the evolvable harness into five modules—Agent Loop, Tool Use, Observation Management, Context Management, and Task Completion Detection—each evolved independently within a restricted scope, then integrated with conflict resolution. Using 2,000 executable evolution tasks disjoint from evaluation benchmarks, it shows consistent gains on TB2.0 and SWE-Bench Verified and transfers across different foundation models.
- Decomposes agent harness into five independently evolving modules with an integration stage
- Contrastive analysis of successful vs failed trajectories identifies recurring behavioral deficiencies
- 2,000 benchmark-disjoint evolution tasks prevent benchmark-specific overfitting
- Consistent gains on TB2.0 and SWE-Bench Verified; transfers across foundation models
- AI models
- ModularRSISWE-Bench Verified
From the original article
1 min read at huggingface.co
Recent work extends recursive self-improvement (RSI) to agent harnesses for long-horizon coding and terminal tasks, enabling agents to improve execution mechanisms from experience. However, generalizable harness RSI remains challenging. First, evolving harnesses on evaluation benchmarks or their subsets makes it difficult to distinguish reusable improvements from benchmark-specific adaptation.
By Siwei Wu, Jincheng Ren, Yizhi Li · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.