ClawLoop: Self-Improving Agents via Harness-Native Training Loop

Hugging Face Daily Papers Papers
llm-agents self-improvement reinforcement-learning post-training harness-environments task-synthesis

Summary

ClawLoop 提出一个无需外部教师模型的自改进训练框架，通过任务合成、训练与诊断反馈三个阶段形成自我维持循环，持续提升基于 harness 环境的 LLM 智能体能力，并在多个基准上验证了效果。

The rapid advancement of Large Language Models (LLMs) has driven their deployment toward general-purpose harness environments, where LLM-based agents operate in sandbox workspaces and interact directly with file systems and shells. While numerous post-training pipelines exist for tool-use agents, their counterparts for harness-based agents remain scarce and are predominantly one-shot and teacher-dependent: a stronger external model synthesizes tasks and collects trajectories, the target agent is trained once, and evaluation results are consumed by human researchers rather than fed back into the loop. Such pipelines are inherently bounded by the capability of the external model and the scope of human prior, leaving no room for self-improvement. To address these limitations, we propose ClawLoop, a framework that enables self-improving without any external teacher model and with minimal prior dependencies. ClawLoop composes three phases into a self-sustaining cycle: Task Synthesis, which generates a diverse pool of training tasks via two complementary construction routines, applies diagnosis-conditioned difficulty growth followed by both agent-free and agent-based task validation; Training, which updates the agent via SFT and/or RL; and Diagnosis and Feedback, which analyzes the agent's rollout behaviors to produce a capability analysis report that will be used as feedback for task synthesis in the next iteration. Using Qwen3.6-27B as the loop model, ClawLoop yields consistent improvements on SkillsBench, Claw-Eval-Live, and ClawGym-Bench. The iteratively refined tasks are further transferred to smaller-scale Qwen3.5-4B and 9B models with gains across all benchmarks, validating data quality independently of the loop model.
Original Article
View Cached Full Text

Cached at: 10/14/26, 06:54 AM

Paper page - ClawLoop: Self-Improving Agents via Harness-Native Training Loop

Source: https://huggingface.co/papers/2610.14118

Abstract

TherapidadvancementofLargeLanguageModels(LLMs)hasdriventheirdeploymenttowardgeneral-purposeharnessenvironments,whereLLM-basedagentsoperateinsandboxworkspacesandinteractdirectlywithfilesystemsandshells.Whilenumerouspost-trainingpipelinesexistfortool-useagents,theircounterpartsforharness-basedagentsremainscarceandarepredominantlyone-shotandteacher-dependent:astrongerexternalmodelsynthesizestasksandcollectstrajectories,thetargetagentistrainedonce,andevaluationresultsareconsumedbyhumanresearchersratherthanfedbackintotheloop.Suchpipelinesareinherentlyboundedbythecapabilityoftheexternalmodelandthescopeofhumanprior,leavingnoroomforself-improvement.Toaddresstheselimitations,weproposeClawLoop,aframeworkthatenablesself-improvingwithoutanyexternalteachermodelandwithminimalpriordependencies.ClawLoopcomposesthreephasesintoaself-sustainingcycle:TaskSynthesis,whichgeneratesadiversepooloftrainingtasksviatwocomplementaryconstructionroutines,appliesdiagnosis-conditioneddifficultygrowthfollowedbybothagent-freeandagent-basedtaskvalidation;Training,whichupdatestheagentviaSFTand/orRL;andDiagnosisandFeedback,whichanalyzestheagent’srolloutbehaviorstoproduceacapabilityanalysisreportthatwillbeusedasfeedbackfortasksynthesisinthenextiteration.UsingQwen3.6-27Bastheloopmodel,ClawLoopyieldsconsistentimprovementsonSkillsBench,Claw-Eval-Live,andClawGym-Bench.Theiterativelyrefinedtasksarefurthertransferredtosmaller-scaleQwen3.5-4Band9Bmodelswithgainsacrossallbenchmarks,validatingdataqualityindependentlyoftheloopmodel.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2610\.14118

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/2610.14118 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.14118 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.14118 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

Recurrent Harness for Agentic Continual Learning

arXiv cs.LG

提出 ReHarness 框架，通过在严格在线流式协议下结合前向读写与后向修订（利用轨迹反馈修订检索到的反思），实现 LLM 智能体的持续学习，在三个基准和三种模型骨干上均优于基线方法。

Self-Harness: Harnesses That Improve Themselves

Hacker News Top

Self-Harness introduces a new paradigm where LLM-based agents iteratively improve their own operating harness by mining model-specific weaknesses, proposing harness modifications, and validating them through regression testing, achieving substantial performance gains on Terminal-Bench-2.0 across multiple base models.

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。