HyperSkill: Self-Evolving LLM Agents via Hypergraph-Structured Skill Memory

2026-08-17

AI Summary

This study addresses the lack of subtask-skill composition modeling and structured maintenance in existing agent memory systems by proposing a hypergraph-based self-evolving experiential memory framework. Introducing a novel dual-node hypergraph structure, this approach jointly optimizes storage, retrieval, and maintenance mechanisms. Through dual-path retrieval, co-occurrence ranking, and quality-weighted propagation, it effectively merges redundant skills and prunes inefficient nodes. Experimental evaluations on xBench, GAIA, and WebWalkerQA demonstrate that the proposed framework significantly outperforms ten baseline models, achieving performance improvements of 11.51% on GAIA and 11.18% on WebWalkerQA. These results substantiate the efficacy of structured memory in enhancing agent capabilities.

Abstract

As agentic tasks grow in complexity, LLM agents increasingly rely on experiential memory to reuse procedural knowledge across tasks. Effective memory design must jointly address what to store, how memory is structured and retrieved, and how memory evolves. Existing systems tackle each only partially: they store trajectories, insights, or workflows as isolated entries, discarding compositional relationships among subtasks and reusable skills; retrieve by flat embedding similarity that ignores relational signals; and maintain memory without leveraging its relational structure. We propose HyperSkill, a hypergraph-based memory framework that jointly improves all three. HyperSkill represents memory as a hypergraph with two node types, subtask steps and reusable skills, where each hyperedge links the subtasks and skills from a single trajectory. Dual-path retrieval queries both subtask and trajectory levels, ranking skills by co-occurrence across retrieved trajectories. Periodic structure-informed maintenance prunes low-utility nodes and merges redundant skills via quality-weighted propagation. Across xBench, GAIA, and WebWalkerQA with GPT-4o and Qwen3-30B-A3B, HyperSkill outperforms ten memory baselines, yielding gains of up to +11.51 on GAIA and +11.18 on WebWalkerQA.

Problem

• LLM Agents
• Experiential Memory
• Skill Reuse
• Memory Retrieval
• Memory Maintenance

Innovation

• Hypergraph-Structured Memory
• Self-Evolving Agents
• Dual-Path Retrieval
• Skill Reuse
• Structure-Informed Maintenance