cs.AINov 2, 2026

From Local Skills to Long-Horizon Web Tasks: Progressive Skill Exploration for LLM Web Agents

Authors: Xueyang Feng, Xiaohe Bo, Xu Chen, Quanyu Dai, Feng Wen

Organizations: Gaoling School of Artificial Intelligence, Renmin University of China · Huawei Technologies Ltd.

Abstract

Large language models (LLMs) are driving autonomous agents toward real-world environments, with the Web emerging as one of the most representative interaction scenarios. However, training Web agents with naive exploration remains challenging: successful trajectories are rare, and the sparse terminal rewards make it difficult for agents to obtain useful feedback from failed trajectories. Inspired by skill acquisition theory, we propose Skill-Composed Progressive Exploration (SCOPE), a short-to-long learning framework for long-horizon Web agents that decomposes complex task exploration into a gradual expansion process from local skills to complete task-solving capability. Specifically, SCOPE iteratively explores around local skills: Skill Discovery extracts reusable local skills and failure experiences from historical rollouts; Skill Orchestration composes discovered skills into high-level skill chains; and Skill Expansion leverages failure experiences aligned with the current skill chain to guide the agent toward exploring new subskills. Through this process, SCOPE progressively constructs task-completing skill chains across rollouts and internalizes them into the policy parameters through agentic on-policy distillation. Extensive experiments on WebArena, WebVoyager, and WebShop show that SCOPE consistently improves exploration dynamics and achieves 3.9% to 18.6% relative gains over the strongest baselines in each setting.

Figures & tables

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. WebXSkill: Skill Learning for Autonomous Web Agents

    Apr 14, 2026Zhaoyang Wang, Qianhui Wu, Xuchao Zhang +12LLM Agent Skill LearningLong-Horizon LLM Agents

  2. Beyond Domains: Reusing Web Skills via Transferable Interaction Patterns

    Jun 16, 2026Shiqi He, Yue Cui, Feijie Wu +5Tool-Augmented Language Model AgentsLLM Agent Skill Learning