Skip to content

SkillEvoReg: Regularizing Agent Skill Evolution Against Overfitting

Sep 2026 · arXiv.org · Vol abs/2609.30861 · 3 citations · 27 references
Computer Science

TL;DR

SkillEvoReg, a general regularization framework for skill evolution inspired by anti-overfitting techniques in neural-network training, is introduced and suggested that explicit regularization is a useful complement to increasingly capable skill updaters.

Abstract

Language-model agents increasingly improve by converting execution experience into reusable external skills. Yet repeated skill updates form a learning process of their own: locally useful edits can accumulate into redundant or task-specific instructions, while new updates can disrupt behavior that previously worked. We study this problem as skill-evolution overfitting and introduce SkillEvoReg, a general regularization framework for skill evolution inspired by anti-overfitting techniques in neural-network training. SkillEvoReg combines training-time skill dropout, which perturbs update generation, and complexity-aware local regularization, which controls unnecessary structural growth, with causal counterexample validation (CCV), which provides targeted behavioral validation of candidate-specific regressions. We instantiate the framework across heterogeneous skill-evolution systems while retaining each system's native skill evolver and task evaluator. Across SkillOpt, SkillEvolBench, and ContinualSkillBench, SkillEvoReg consistently controls skill-state growth while preserving competitive downstream capability, improves several transfer and later-stage evolution outcomes, and identifies update-level regressions that structural metrics alone cannot reveal. These results suggest that explicit regularization is a useful complement to increasingly capable skill updaters.

View source

Similar papers

Preprint Aug 2026

Learning Globally Reusable Skills for Coding Agents

This work proposes GSE, a globalized skill evolution framework that jointly optimizes skill compatibility and skill generalization, and maintains a Skill Relation Graph (SRG) that explicitly models and co-evolves inter-skill relationships.

Chen Yang, Jiashuo Tian, Zi-Qi Wang et al. · 3 citations · ⚡1
#artificial intelligence Preprint Sep 2026

SkillAdam: Stable and Efficient Skill Evolution for Agents

SkillAdam, an Adam-inspired framework for optimizing discrete and non-differentiable skill documents, is introduced, an Adam-inspired framework for optimizing discrete and non-differentiable skill documents with substantially fewer optimization iterations and lower cost than prior methods.

Gao-Yuan Li, Mei-Hao Fan, Yi-Zhe Liu et al. · 5 citations · ⚡3
Preprint Aug 2026

SkillCommit: Evolving Agent Skills through Behaviorally Validated Scope Expansion

SkillCommit is an online skill evolution framework that continuously transforms experience into a hierarchical library of reusable skills, enabling cross-model experience transfer and consistently improves agent performance across diverse domains.

Yu He, Wei-Kai Yang · 4 citations
Preprint Aug 2026

DIVE: Unlocking Self-Improvement in Frozen Language Models Through Diversity-Driven Skill Evolution

DIVE is introduced, a diversity-driven framework that enables frozen LLMs to improve by evolving persistent natural-language skills from task experience and verifier feedback and consistently outperforms existing reasoning methods, prompt-optimization approaches, skill-development frameworks, and memory-based baselines...

Siheng Xiong, Ali Payani, Oguzhan Gungordu et al. · 3 citations
#artificial intelligence Preprint Sep 2026

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

COBRA-Skills is introduced, an efficient framework that formulates skill optimization as budgeted sequential optimization over a dynamically evolving candidate space and remains robust to changes in the agent harness and performs effectively when the target model itself is used for skill generation and refinement.

Ping-Chen Lu, Xiang-Yi Wang, Xiang Li et al. · 2 citations

SkillFocus: Evolving Agent Skills via Capability Decomposition

Across four benchmarks spanning heterogeneous tasks, SkillFocus achieves the best held-out accuracy on all four, outperforming the strongest competing result by 5.7 points on average while using 24\% fewer evolution tokens on average than the closest iterative baseline.

Ning Wang, Z. Gong, Bing-Dong Li et al. · 0 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.