paper-with-me

Papers

SkillFlow:Benchmarking Lifelong Skill Discovery and Evolution for Autonomous Agents

2026-04-19 · Ziao Zhang, Kou Shi, Shiting Huang, Avery Nie, Yu Zeng, Yiming Zhao, Zhen Fang, Qishen Su, Haibo Qiu, Wei Yang, Qingnan Ren, Shun Zou, Wenxuan Huang, Lin Chen, Zehui Chen, Feng Zhao arxiv

As the capability frontier of autonomous agents continues to expand, they are increasingly able to complete specialized tasks through plug-and-play external skills. Yet current benchmarks mostly test whether models can use provided skills, leaving open whether they can discover skills from experience, repair them after failure, and maintain a coherent library over time. We introduce SkillFlow, a benchmark of 166 tasks across 20 families in which task construction within each family follows a Domain-Agnostic Execution Flow (DAEF) that defines an agent workflow framework, allowing these tasks to share a consistent workflow. Agents are evaluated under an Agentic Lifelong Learning protocol in which they begin without skills, solve tasks sequentially within each family, externalize lessons through trajectory- and rubric-driven skill patches, and carry the updated library forward. Experiments reveal a substantial capability gap. For Claude Opus 4.6, lifelong skill evolution improves task success from 62.65% to 71.08% (+8.43 points). However, high skill usage does not necessarily imply high utility: Kimi K2.5 gains only +0.60 points despite 66.87% skill usage, while Qwen-Coder-Next reaches only a 44.58% task completion rate and still regresses relative to the vanilla setting. SkillFlow contributes a structured testbed for this direction and an in-depth empirical analysis of skill discovery, patching, transfer, and their failure modes under lifelong evaluation.

📄 PDF Abstract BibTeX arXiv:2604.17308

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

SkillFlow: Flow-Driven Recursive Skill Evolution for Agentic Orchestration

2026-05-13 · Mingda Zhang, Tiesunlong Shen, Haoran Luo, Wenjin Liu 외 arxiv

In recent years, a variety of powerful LLM-based agentic systems have been applied to automate complex tasks through task orchestration. However, existing orchestration methods still face key challenges, including strate…

Mathematical ReasoningQuestion AnsweringCode GenerationDecision Making

SkillFlow: Efficient Skill and Code Transfer Through Communication in Adapting AI Agents

2025-04-08 · Pagkratios Tagkopoulos, Fangzhou Li, Ilias Tagkopoulos

AI agents are autonomous systems that can execute specific tasks based on predefined programming. Here, we present SkillFlow, a modular, technology-agnostic framework that allows agents to expand their functionality in a…

Scheduling

Learning While Acting: A Skill-Enhanced Test-Time Co-Evolution Framework for Online Lifelong Learning Agents

2026-06-03 · Bo Mao, Jie Zhou, Yutao Yang, Xin Li 외 arxiv

Lifelong learning is essential for Large Language Model (LLM) agents operating in dynamic, interactive environments. However, existing lifelong learning agents for long-horizon tasks typically depend on discrete skill or…

Reinforcement Learning

Reset-Free Lifelong Learning with Skill-Space Planning

2020-12-07 · ICLR 2021 1 · Kevin Lu, Aditya Grover, Pieter Abbeel, Igor Mordatch

The objective of lifelong reinforcement learning (RL) is to optimize agents which can continuously adapt and interact in changing environments. However, current RL approaches fail drastically when environments are non-st…

Lifelong learningMuJoCoReinforcement Learning (RL)

AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

2026-03-01 · Yutao Yang, Junsong Li, Qianjun Pan, Bihao Zhan 외 arxiv

In practical LLM applications, users repeatedly express stable preferences and requirements, such as reducing hallucinations, following institutional writing conventions, or avoiding overly technical wording, yet such in…