Harmony Harness：Agent Harness 的多目标自进化框架

View PDF HTML (experimental)

Abstract:Recent work on self-evolving LLM agents treats the agent harness as an optimization target that can be iteratively improved through execution feedback. Existing approaches, however, largely optimize a single objective and evolve the harness as a monolithic artifact, limiting their ability to handle competing goals and to separate transferable design from environment-specific adaptation. We introduce Harmony Harness, a framework for multi-objective, compositional harness evolution. Harmony Harness formulates harness evolution as a multi-objective optimization problem and uses decomposition-based evolutionary search to maintain solutions across different regions of the Pareto frontier, avoiding reliance on a hand-designed structural diversity metric for repository-scale harnesses. In parallel, we introduce factorized evolution operators that separately refine general-purpose harness mechanisms and benchmark-specific adaptations, together with a dynamic composition workflow that integrates them at inference time. Across SecRepoBench and BaxBench, Harmony Harness consistently outperforms strong harness-evolution baselines, with gains exceeding 10% on both benchmarks. Our results show that explicitly modeling both objective trade-offs and harness structure yields substantially stronger and more transferable agent evolution than existing single-objective, monolithic approaches.

Subjects:	Artificial Intelligence (cs.AI)
Cite as:	arXiv:2610.15150 [cs.AI]
	(or arXiv:2610.15150v1 [cs.AI] for this version)
	https://doi.org/10.48550/arXiv.2610.15150 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Weichen Yu [view email]
[v1] Sat, 10 Oct 2026 20:06:59 UTC (541 KB)