paper-with-me

홈 › Papers

From Interaction Traces to Persistent Skills: Online Evolution for Computer-Use Agents

2026-09-04 · Longtao Hu, Xiao Liang, Linchao Zhu arxiv

Computer-use agents can execute increasingly complex tasks in graphical interfaces, but their interaction experience is typically transient: procedural knowledge acquired from one rollout is not systematically retained, refined, and reused in later tasks. Existing skill libraries provide external procedural knowledge, yet their incremental value over the same agent operating without skills, as well as their longitudinal dynamics under repeated interaction, remain insufficiently characterized. We present an online skill-evolution framework that converts interaction trajectories and evaluator feedback into a persistent, versioned library of reusable procedures. Each iteration executes against a frozen library snapshot, and evidence-guided skill updates become available in subsequent iterations without changing model parameters. We compare the full evolving-library system with a configuration-matched empty-library control across four OSWorld application domains under the same fixed action-generation and GUI-grounding stack, task sets, and iteration horizons. Following a five-iteration empty-library warm-up, Full attains a higher post-warm-up mean evaluator score in all four observed domain runs, with mean differences ranging from 5.7 to 18.6 percentage points and domain-dependent temporal stability. In GIMP, provenance-aware analysis reveals retrieval across task-of-origin boundaries and revision churn, where repeated accepted edits fail to recover the originating task. These findings characterize evolving skill libraries as auditable, shared procedural memory that can improve a fixed computer-use stack, while showing that their benefits are conditional and repeated revision does not guarantee recovery. Code is released at https://github.com/LongtaoHu/Skill-Evo4GUI.

📄 PDF Abstract BibTeX arXiv:2609.04869

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

RoboFoundry: System-as-Policy Evolution for Self-Learning Embodied Agents

2026-09-26 · Jingsong Liang, Shuhao Liao, Shizhe Zhang, Diyuan Hou 외 hf

A foundation model should not act in isolation as an embodied agent. Yet, existing methods often optimize individual components of the agent stack, such as memory, context, skills, or action interfaces, rather than treat…

Online Training of Large Language Models: Learn while chatting

2024-03-04 · Juhao Liang, Ziwei Wang, Zhuoheng Ma, Jianquan Li 외

Large Language Models(LLMs) have dramatically revolutionized the field of Natural Language Processing(NLP), offering remarkable capabilities that have garnered widespread usage. However, existing interaction paradigms be…

WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution

2026-08-27 · Liyan Tang, Cyrus Rashtchian, Chun-Sung Ferng, Andrew Tomkins 외 hf

Agent skills package specialized knowledge and workflows into reusable resources that extend AI agent capabilities. Recent work automatically discovers such skills from agent experience, which enables agents to progressi…

EvoSkill Injection: Red-Teaming Autonomous Skill Generation and Evolution in Self-Evolving Agents

2026-08-31 · Doyun Kim, Chanwoo Kim, Sugyeong Eo, Yeo-Chan Yoon 외 arxiv

LLM-based agent systems increasingly adopt skill-based architectures to reduce repetitive reasoning costs and improve stable, efficient task execution. Recent studies propose self-evolving agents that autonomously genera…

SkillSmith: Co-Evolving Skills and Tools for Self-Improving Agent Systems

2026-05-31 · Yangbo Wei, Zhen Huang, Shaoqiang Lu, Junhong Qian 외 arxiv

Recent self-evolving agents have shown that skills can be discovered, refined, and accumulated through execution. However, existing skill-evolution frameworks typically assume a fixed tool layer and evaluate each skill i…