Harness^2: Recursive Agent Harnessing for an Open World

Hugging Face Daily Papers Papers
llm-agents recursive-self-improvement agent-harness benchmarks open-world-tasks

Summary

Harness² 提出了一种即插即用的递归智能体框架（Recursive Agent Harnessing），通过任务级与框架级两层递归改进，在无 ground-truth 验证器的开放世界场景中自动优化智能体执行框架，并在多个专业基准上显著提升了 GLM-5 与 Gemini 3.5 Flash 的表现。

LLM agents' capabilities depend not only on their underlying models but also on their execution harnesses. Recursive self-improvement (RSI) offers a promising approach to automatically evolve these harnesses. However, deploying RSI in open-world settings still faces two critical challenges. First, open-world tasks are typically unseen during the harness design phase and novel at test time, leaving no ground-truth verifiers to reliably guide online harness edits. Second, the RSI process remains highly task-specific and fails to generalize to novel scenarios. We introduce Harness^2, a plug-and-play, recursive agent harnessing framework that addresses these challenges through two connected levels of recursion. Task-level recursion refines the current task's harness using a contrastive Propose-Probe-Compose step that reliably guides edits without ground-truth verifiers, and can be scaled further via parallel or sequential recursion. Harnessing-level recursion then extracts experience from these edits to update a continual improvement harness, enabling the agent to better adapt to future tasks. Across three professional benchmarks and four model-harness configurations, a single step consistently improves GLM-5 and Gemini 3.5 Flash over their base harnesses, yielding gains of up to 11.5 points on WorkBuddy-Bench, 4.5 points on JobBench, and 11.8 percentage points on LAB, with further gains as the task-level recursion budget increases.
Original Article
View Cached Full Text

Cached at: 10/14/26, 02:49 AM

Paper page - Harness^2: Recursive Agent Harnessing for an Open World

Source: https://huggingface.co/papers/2610.14224

Abstract

LLMagents’capabilitiesdependnotonlyontheirunderlyingmodelsbutalsoontheirexecutionharnesses.Recursiveself-improvement(RSI)offersapromisingapproachtoautomaticallyevolvetheseharnesses.However,deployingRSIinopen-worldsettingsstillfacestwocriticalchallenges.First,open-worldtasksaretypicallyunseenduringtheharnessdesignphaseandnovelattesttime,leavingnoground-truthverifierstoreliablyguideonlineharnessedits.Second,theRSIprocessremainshighlytask-specificandfailstogeneralizetonovelscenarios.WeintroduceHarness^2,aplug-and-play,recursiveagentharnessingframeworkthataddressesthesechallengesthroughtwoconnectedlevelsofrecursion.Task-levelrecursionrefinesthecurrenttask’sharnessusingacontrastivePropose-Probe-Composestepthatreliablyguideseditswithoutground-truthverifiers,andcanbescaledfurtherviaparallelorsequentialrecursion.Harnessing-levelrecursionthenextractsexperiencefromtheseeditstoupdateacontinualimprovementharness,enablingtheagenttobetteradapttofuturetasks.Acrossthreeprofessionalbenchmarksandfourmodel-harnessconfigurations,asinglestepconsistentlyimprovesGLM-5andGemini3.5Flashovertheirbaseharnesses,yieldinggainsofupto11.5pointsonWorkBuddy-Bench,4.5pointsonJobBench,and11.8percentagepointsonLAB,withfurthergainsasthetask-levelrecursionbudgetincreases.

View arXiv pageView PDFProject pageGitHubAdd to collection

Get this paper in your agent:

hf papers read 2610\.14224

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.14224 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.14224 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.14224 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Google Harness² for Recursive Self Improvement (GitHub Repo)

TLDR AI

Google Research 开源 Harness²，一个实现递归式智能体系统优化（recursive agent harnessing）的即插即用框架，可在模型权重固定的情况下通过 Propose–Probe–Compose 循环持续改进 agent harness，在多个专业基准上带来最高 11.8 个百分点的性能提升。