Paper page - Evolving to Match Capabilities: Improving Harness Self-Evolution via Skill Adaptation

Source: https://huggingface.co/papers/2610.33024

Abstract

Harness self-evolution improves LLM agents via skill adaptation guided by a meta-skill that reformulates guidance to match executor capabilities, boosting weak models without parameter updates.

Harness self-evolutionoffers a practical route forLLM agentsto improve from execution experience without updating model parameters. Its effectiveness depends on whether models can successfully act on the guidance provided by their harnesses. However, we find that harness benefits are not uniform across model capabilities: skills that improve stronger models can provide little benefit or even degrade performance for weaker ones. Across seven models and three benchmarks, we identify 103 task-skill pairs where skills improve GPT-5.6-luna by 31.1 points but degrade Gemma-4-E4B by 8.8 points. In 86% of failed weak-model runs, the violated requirement is already stated in the skill. This motivatesskill adaptation: reformulating existing skills to match the executor’s capabilities and make their guidance easier to execute. To achieve this, we diagnose failures intool execution,context management,instruction following, andverification, and distill these diagnoses into a fixedmeta-skillthat guides adaptation duringharness self-evolution. Guided by themeta-skill, a capable external rewriter, Kimi-K3, adapts skills without execution evidence, improving the three weak executors by 5.1-14.1 points over the original skills. We then incorporate this adaptation into a five-roundself-evolution loop, where each model revises its own skill from execution trajectories and verifier feedback. Across seven executors, from Gemma-4-E4B to Kimi-K2.5, themeta-skillconsistently improves final scores by 2.0-5.1 points over the same loop without themeta-skill. Our findings provide practical design guidance for developing more effectiveharness self-evolutionmethods.

View arXiv pageView PDFProject pageGitHub2Add to collection

Get this paper in your agent:

hf papers read 2610\.33024

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/2610.33024 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.33024 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.33024 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.