Beyond Single-Shot: Multi-step Tool Retrieval via Query Planning
This addresses the challenge of effective tool retrieval for LLM agents in dynamic environments, representing a novel method for a known bottleneck rather than a foundational advancement.
The paper tackles the problem of LLM agents struggling with complex tool retrieval due to semantic gaps and limited embedding capacity, proposing TOOLQP, a framework that models retrieval as iterative query planning, which achieves state-of-the-art performance with superior zero-shot generalization and robustness.
LLM agents operating over massive, dynamic tool libraries rely on effective retrieval, yet standard single-shot dense retrievers struggle with complex requests. These failures primarily stem from the disconnect between abstract user goals and technical documentation, and the limited capacity of fixed-size embeddings to model combinatorial tool compositions. To address these challenges, we propose TOOLQP, a lightweight framework that models retrieval as iterative query planning. Instead of single-shot matching, TOOLQP decomposes instructions into sub-tasks and dynamically generates queries to interact with the retriever, effectively bridging the semantic gap by targeting the specific sub-tasks required for composition. We train TOOLQP using synthetic query trajectories followed by optimization via Reinforcement Learning with Verifiable Rewards (RLVR). Experiments demonstrate that TOOLQP achieves state-of-the-art performance, exhibiting superior zero-shot generalization, robustness across diverse retrievers, and significant improvements in downstream agentic execution.