Metraipolis › archive › dc_256a501ed980

ToolSearcher: Optimizing Tool Selection at Scale via Reinforcement Learning — arXiv:2609.30906

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.30906
License (as archived)
CC0-1.0
Archived
2026-10-08T02:04:02.361Z
SHA-256 of the archived text
cace5f77da66426f46049a981f78bc371261eac33c1ea380c71b6a7e4b22f305
Stable URL of this page
https://metraipolis.com/documento/dc_256a501ed980

How to cite

ToolSearcher: Optimizing Tool Selection at Scale via Reinforcement Learning — arXiv:2609.30906 — arXiv (en), https://arxiv.org/abs/2609.30906 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_256a501ed980 (Metraipolis archive dc_256a501ed980, sha256 cace5f77da66426f46049a981f78bc371261eac33c1ea380c71b6a7e4b22f305)
BibTeX
@misc{dc256a501ed980,
  title = {ToolSearcher: Optimizing Tool Selection at Scale via Reinforcement Learning — arXiv:2609.30906},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.30906},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_256a501ed980; sha256 cace5f77da66426f46049a981f78bc371261eac33c1ea380c71b6a7e4b22f305},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30906. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) excel at natural language processing but struggle to interact with external environments. Tool learning provides a promising way to extend LLMs into actionable agents, where tool selection is a critical prerequisite for successful tool use. Existing work often assumes a small or predefined set of tools, leaving large-scale tool selection underexplored. Real-world repositories contain a vast and diverse array of tools, making it difficult for LLMs to effectively search, distinguish, and compose tools under context-length constraints. We identify large-scale tool selection as a new challenge for agentic reinforcement learning, highlighting that existing RL methods for knowledge-based question answering are inadequate for selecting tools while considering compatibility. To address this challenge, we propose ToolSearcher, a novel RL framework for effective multi-turn search and fine-grained optimization in large-scale tool selection. Specifically, we introduce category-constrained tool discrimination to improve the model's ability to distinguish functionally similar tools, event-level search modeling to explicitly optimize the discovery of target tools during multi-turn search, and trajectory-aligned credit allocation to provide fine-grained reward signals for different stages of the search-selection process. Extensive experiments on large-scale tool selection benchmarks demonstrate that ToolSearcher consistently outperforms a set of strong baselines in challenging settings involving iterative search and complex tool composition.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.30906 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T01:38:49.446Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).