Hugging Face Daily PapersSiwei Wu, Jincheng Ren, Yizhi LiSep 14, 20261 min readpaperadvanced

ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement

Summary

ModularRSI proposes a contrastive, modular approach to evolve agent harnesses by independently updating five functional modules and aggregating evidence across 2,000 benchmark‑disjoint tasks. Experiments on TB2.0 and SWE‑Bench Verified show consistent gains on unseen tasks and cross‑model transfer.

The harness is decomposed into Agent Loop, Tool Use, Observation Management, Context Management, and Task Completion Detection, each evolved within a restricted scope.
Contrastive analysis of successful vs. failed trajectories across 2k external tasks isolates recurring behavioral deficiencies.
Benchmark‑disjoint evolution prevents overfitting to evaluation suites, improving generalization.
Evolved harnesses achieve measurable performance gains on unseen in‑domain and cross‑domain tasks in TB2.0 and SWE‑Bench Verified.

Engineers building LLM‑driven coding agents need a systematic way to improve their control harnesses without overfitting, and ModularRSI offers a modular, evidence‑driven solution.

7/10

llm architecture performance ml infra systems programming

Related reading

The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement arxiv.org
The paper defines recursive self‑improvement (RSI) for AI, introduces the Headroom‑Closed Index to expose current LLM limits, and proposes a staged roadmap toward full RSI. It surveys application scenarios and outlines practical challenges.
Hugging Face Daily Papersarxiv.orgYi Duan, Ying Liu, Zirui Tang12d1 minpaperHN3
HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harness arxiv.org
HarnessVLN introduces a zero‑shot, training‑free embodied navigation framework that wraps a multimodal LLM in an "Agent Harness" – a tool‑based protocol that validates planner actions against spatial evidence, tracks progress with hierarchical event memory, and maintains a persistent spatiotemporal graph for recovery. The system works for instruction‑following and object‑goal tasks, achieving 60.…
Hugging Face Daily Papersarxiv.orgYang Chen, Lirong Che, Zhenyu Huang8d1 minpaper
COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization arxiv.org
COBRA‑Skills uses a contextual‑bandit loop to selectively evaluate and evolve LLM agent skills, achieving better performance with roughly half the evaluation cost of prior methods. The framework works with limited examples and remains robust across different agent setups and even when the target model creates its own skills.
Hugging Face Daily Papersarxiv.orgPingchen Lu, Xiangyi Wang, Xiang Li12d1 minpaper
Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement arxiv.org
The paper introduces Generalized Agent Iteration (GAI), a formal framework that unifies classical iterative policy improvement (GPI) and recursive self‑improvement (RSI). GAI treats an agent as a set of modifiable components and models learning as a loop of evaluation and improvement. Two binary “dials”—whether the improvement mechanism is internal to the agent and whether the evaluation standard…
Hugging Face Daily Papersarxiv.orgHongyao Tang, Yi Ma, Pengyi Li11d1 minpaper
An Empirical Study of Harness Design for Coding Agents arxiv.org
This empirical study dissects coding agent harness components like planning, action space, and context management. It finds context management is critical for tight windows, rule-based elision is efficient, and planning shifts from accuracy to cost savings for stronger models.
Hugging Face Daily Papersarxiv.orgRun-Ze Fan, Zihao Zhang, Simin Ma5d2 minpaperHN12621
Presentation: The Agent Harness: Control Planes, Invariants, and Approval Boundaries for Production AI Agents infoq.com
Production AI agents often fail silently due to issues beyond model hallucination, such as state inconsistencies. This talk outlines principles for building reliable agent harnesses, focusing on explicit state ownership, ordered mutations, and verifiable actions, using real-world case studies.
InfoQinfoq.comVinoth Govindarajan18h24 mintalk

ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement

Summary

Related reading

The Last AI Built by Humans: Toward Genuine Recursive Self-Improvementarxiv.org

HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harnessarxiv.org

COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimizationarxiv.org

Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvementarxiv.org

An Empirical Study of Harness Design for Coding Agentsarxiv.org

Presentation: The Agent Harness: Control Planes, Invariants, and Approval Boundaries for Production AI Agentsinfoq.com

The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement arxiv.org

HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harness arxiv.org

COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization arxiv.org

Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement arxiv.org

An Empirical Study of Harness Design for Coding Agents arxiv.org

Presentation: The Agent Harness: Control Planes, Invariants, and Approval Boundaries for Production AI Agents infoq.com