{
  "annotation_id": "0a4536e3-2095-4121-9c37-e96254adab1c",
  "date_created": "2026-02-17T05:53:11.954000Z",
  "date_modified": "2026-02-17T05:53:11.954000Z",
  "file_hash": "afed5d84eb30666ec54c65ed9886551692282c1389efd5c6709b3f92cdb74b38",
  "private": false,
  "record": {
    "abstract": "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.",
    "arxiv_id": "2601.07782",
    "authors": [
      "Wei Fang",
      "James Glass"
    ],
    "categories": [
      "cs.CL",
      "cs.AI",
      "cs.IR"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "title": "Beyond Single-Shot: Multi-step Tool Retrieval via Query Planning",
    "url": "https://arxiv.org/abs/2601.07782",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "e9823d68-f5d6-47fc-9cea-242c94c4ef43",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}