AlignEvoSkill: Towards Knowledge-Aware and Task-Aligned Agent Skill Evolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Dingzirui, Zhang, Xuanliang, Xu, Keyan, Zhu, Qingfu, Che, Wanxiang, Deng, Yang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911717027479552
author Wang, Dingzirui
Zhang, Xuanliang
Xu, Keyan
Zhu, Qingfu
Che, Wanxiang
Deng, Yang
author_facet Wang, Dingzirui
Zhang, Xuanliang
Xu, Keyan
Zhu, Qingfu
Che, Wanxiang
Deng, Yang
contents Reusable skills play a key role in improving LLM-based agents, but existing skill-evolution methods often fail to ensure that evolved skills both cover the knowledge required by the task and remain aligned with the target task. As a result, evolved skills could be incomplete or irrelevant. To address this limitation, we propose AlignEvoSkill, a skill-evolution framework that jointly models knowledge coverage and task alignment. Given failed task trajectories, AlignEvoSkill first identifies task-relevant knowledge tags, retrieves complementary prior skills, and adapts them into candidate skills that address missing knowledge. It then selects high-quality candidates using a joint filtering criterion based on knowledge-coverage and task-alignment scores. Experiments on 3 benchmarks with4 LLM backbones show a 34.7% relative gain of AlignEvoSkill over the non-evolution baseline and achieves a new SOTA in skill evolution with lower cost.
format Preprint
id arxiv_https___arxiv_org_abs_2506_23149
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AlignEvoSkill: Towards Knowledge-Aware and Task-Aligned Agent Skill Evolution
Wang, Dingzirui
Zhang, Xuanliang
Xu, Keyan
Zhu, Qingfu
Che, Wanxiang
Deng, Yang
Computation and Language
Reusable skills play a key role in improving LLM-based agents, but existing skill-evolution methods often fail to ensure that evolved skills both cover the knowledge required by the task and remain aligned with the target task. As a result, evolved skills could be incomplete or irrelevant. To address this limitation, we propose AlignEvoSkill, a skill-evolution framework that jointly models knowledge coverage and task alignment. Given failed task trajectories, AlignEvoSkill first identifies task-relevant knowledge tags, retrieves complementary prior skills, and adapts them into candidate skills that address missing knowledge. It then selects high-quality candidates using a joint filtering criterion based on knowledge-coverage and task-alignment scores. Experiments on 3 benchmarks with4 LLM backbones show a 34.7% relative gain of AlignEvoSkill over the non-evolution baseline and achieves a new SOTA in skill evolution with lower cost.
title AlignEvoSkill: Towards Knowledge-Aware and Task-Aligned Agent Skill Evolution
topic Computation and Language
url https://arxiv.org/abs/2506.23149