Metraipolis › archive › dc_7ff2e944308b
COMPASS: Routing via Capability-Oriented Model Profiling with Adaptive Search, Selection, and Tool Use — arXiv:2610.27822
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.27822
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-21T07:06:52.590Z
- SHA-256 of the archived text
df8454aee0a45c66ef535c4cac8ad5c3cf66fa9045d864b9d2c1fea816d06d1e- Stable URL of this page
https://metraipolis.com/documento/dc_7ff2e944308b
How to cite
COMPASS: Routing via Capability-Oriented Model Profiling with Adaptive Search, Selection, and Tool Use — arXiv:2610.27822 — arXiv (en), https://arxiv.org/abs/2610.27822 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7ff2e944308b (Metraipolis archive dc_7ff2e944308b, sha256 df8454aee0a45c66ef535c4cac8ad5c3cf66fa9045d864b9d2c1fea816d06d1e)
BibTeX
@misc{dc7ff2e944308b,
title = {COMPASS: Routing via Capability-Oriented Model Profiling with Adaptive Search, Selection, and Tool Use — arXiv:2610.27822},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.27822},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7ff2e944308b; sha256 df8454aee0a45c66ef535c4cac8ad5c3cf66fa9045d864b9d2c1fea816d06d1e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27822. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
LLM routing aims to select a capable model at low cost, but specialized routers can require updates as models and tasks change. We propose COMPASS, an agentic router in which an LLM follows a routing recipe, gathers performance and cost evidence through tools, and optionally inspects candidate answers before selecting a model. Our evaluations on LLMRouterBench and DeepSWE show that agentic routing can improve accuracy and lower model-call cost relative to specialized and trained routing baselines, without routing-specific training. On LLMRouterBench, COMPASS reaches 69.85% accuracy at \$139.57, compared with RLCascadeRouter's 69.14% at \$180.22. On DeepSWE under fixed five-fold cross-validation, it reaches 78.1% at \$215, compared with 74.1% at \$500 for the best single model. COMPASS also supports routing across benchmarks and adapts to an expanded model pool by reading new evaluation records, without changing the router's weights or prompt. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.27822 — 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 2026-10-21T07:05:43.449Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).