Metraipolis › archive › dc_ea356017d838
TRACER: Trajectory-Aware Tool Retrieval for LLM Agents — arXiv:2610.21267
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.21267
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T09:19:30.325Z
- SHA-256 of the archived text
d43b7cac825a0acf7d8fae6b80f572b1245e1b13c38ce221dae1a0f2fc39efdb- Stable URL of this page
https://metraipolis.com/documento/dc_ea356017d838
How to cite
TRACER: Trajectory-Aware Tool Retrieval for LLM Agents — arXiv:2610.21267 — arXiv (en), https://arxiv.org/abs/2610.21267 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ea356017d838 (Metraipolis archive dc_ea356017d838, sha256 d43b7cac825a0acf7d8fae6b80f572b1245e1b13c38ce221dae1a0f2fc39efdb)
BibTeX
@misc{dcea356017d838,
title = {TRACER: Trajectory-Aware Tool Retrieval for LLM Agents — arXiv:2610.21267},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.21267},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ea356017d838; sha256 d43b7cac825a0acf7d8fae6b80f572b1245e1b13c38ce221dae1a0f2fc39efdb},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21267. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Modern agents select tools from registries, but exposing entire registries incurs high token costs and can impair selection as they grow. We present TRACER, a retrieval front-end that indexes observed tool-call sequences in a trie and ranks continuations using a learned blend of empirical and contextual priors. On a mixed retail-and-telecom $\tau^2$ environment, TRACER improves GPT-5.4's tool-selection accuracy by 4.7 points while reducing input tokens per step by 80%, exposing just 5 of 128 tools. In autonomous full-turn execution, it matches GPT-5.4's full-registry task success with 61% fewer input tokens per step. Across six environments and four models, TRACER improves tool-selection accuracy over full-registry selection in 21 of 24 model-environment pairs, with average gains of 6 to 10 points; the three exceptions trail by at most 2.2. We analyze failures in both regimes to identify directions for future work. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Information Retrieval (cs.IR). Pagina del lavoro: https://arxiv.org/abs/2610.21267 — 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 2026-10-16T08:44:17.931Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).