HiSkill: Empowering LLM Agents with Hierarchical Skill Graphs
Source: https://arxiv.org/abs/2607.25853
arXiv: 2607.25853
Authors: Yu Hao, Jinxuan Cai, Qi Zhang, Yawen Li, Zhiqiang Zhang, Chuan Shi, Cheng Yang
Published: 2026-07-28
Categories: cs.AI
PDF: https://arxiv.org/pdf/2607.25853v1
Abstract (en)
Skills have become an important abstraction for enabling large language model (LLM) agents to reuse past experience in long-horizon interactive tasks. However, existing trajectory-to-skill methods often produce flat collections of high-level textual skills that are stored and retrieved independently, leaving skill relations underutilized and maintaining a gap between high-level skills and executable actions. In this paper, we propose HiSkill, a hierarchical skill graph framework that organizes interaction trajectories into a directed graph with skill nodes, AtomicOp nodes, and typed edges. Specifically, the graph connects reusable high-level skills with executable action templates, while also capturing decomposition, temporal transition, compatibility, support, and recovery relations among them. At inference time, HiSkill retrieves a compact task-relevant subgraph and performs subgraph-guided task execution, where a symbolic task state, an active skill, and the retrieved subgraph guide the LLM agent to switch skills, select AtomicOps, and ground executable actions iteratively. Experiments on three interactive environments show that HiSkill outperforms state-of-the-art baselines while reducing inference token consumption, demonstrating the effectiveness of bridging high-level skills and executable action grounding through a hierarchical skill graph. Our data and code is available at https://github.com/BUPT-GAMMA/HiSkill.
Summary (zh)
HiSkill 针对 trajectory-to-skill 常见缺陷——扁平文本技能库、技能关系缺失、高层描述与可执行动作断层——提出层次技能图：节点含 skill 与 AtomicOp，边覆盖分解、时序迁移、兼容、支持与恢复。推理时检索任务相关子图，维护 symbolic task state 与 active skill，迭代选择 AtomicOp 并 grounding 为可执行动作。三个交互环境上优于 SOTA，并降低 inference token；代码与数据开源在 BUPT-GAMMA/HiSkill。
One-liner
技能不能只是文案清单：HiSkill 用层次技能图把高层技能接到 AtomicOp，并显式建模分解与恢复。
Obsidian evidence
OpenClaw定时任务/论文流水线/2026-07-29-论文流水线.md- opencli:
arxiv paper 2607.25853 -f json - Run date: 2026-07-29