HyperAgent: Planning and Acting over Tool-Schema Hypergraphs for Tool-Use LLM Agents

Short summary

HyperAgent introduces a Tool-Schema Hypergraph framework where tools are modeled as hyperedges connecting input and output schema nodes, enabling more reliable planning for LLM agents. It extracts task-relevant context graphs, builds schema-aware Task DAGs, and dynamically resolves subtasks through deficit-oriented expansion that retrieves supporting tools based on current state. Experiments on AppWorld show improved task completion with fewer redundant API calls, LLM interactions, and tokens compared to baselines.

• •Models tool relations as a directed hypergraph connecting input/output schemas
• •Dynamic planning via deficit-oriented expansion retrieves supporting tools at runtime
• •Outperforms baselines on AppWorld with fewer API calls and lower token consumption

Generated with AI, which can make mistakes.