cs.CLNov 4, 2026

EvoGuard: Mitigating Degradation in Online Skill Evolution

Authors: Yanli Wang, Zichen Bai, Tianyue Jiang, Kunming Shao, Junhao Shi, Xueqin Liu, Jinguo Sun, Muye Li, 

Organizations: Sun Yat-sen University · The University of Melbourne · JD Tech Genius Team · The Hong Kong University of Science and Technology

Abstract

Skills let language agents reuse experience across tasks as natural-language documents. Recent work evolves skills from interaction, but existing methods have limitations when skills evolve online. Prior methods update a skill offline from a batch of trajectories collected over many tasks. Some then adopt the new skill without any check, which can degrade the skill over time. Others keep it only if it improves performance on a held-out validation set, which is too slow and costly to run after every update. We introduce EvoGuard, a pairwise skill selector that mitigates skill degradation at low latency and cost. We first run the agent iteratively to build pools of skills and score each skill on a private test set to obtain utility labels. We pair the skills and use rejection sampling to keep only pairs with a clear utility gap. The selector is trained on the logits of two label tokens to pick the better skill. At deployment it decides in a single forward pass, which is significantly faster than evaluation on the private test set. Compared with adopting every new skill, EvoGuard improves success by 11.0% on ALFWorld, 5.9% on ScienceWorld, and 2.7% on LiveCodeBench.

Figures & tables

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. SkillEvoReg: Regularizing Agent Skill Evolution Against Overfitting

    Sep 25, 2026Guanyu Nie, Fangzhou Zhu, Shixiong Kai +3LLM Agent Skill LearningAgent Skill Learning

  2. SkillEvolver: Skill Learning as a Meta-Skill

    May 11, 2026Genrui Zhang, Erle Zhu, Jinfeng Zhou +2LLM Agent Self-ImprovementLLM Agent Skill Learning

  3. Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning

    May 7, 2026Yaorui Shi, Yuxin Chen, Zhengxi Lu +6LLM Agent Skill LearningLLM Agent Skill Retrieval