Metraipolis › archive › dc_36479aca0b52

Chance-Constrained Personalized LLM Routing with Online Preference Learning — arXiv:2610.22695

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.22695
License (as archived)
CC0-1.0
Archived
2026-10-19T08:48:27.194Z
SHA-256 of the archived text
bc2bbf0cbe36c3a8adf5f6a92dd56507b8235576d1f989ed8dfb0a26b96b5980
Stable URL of this page
https://metraipolis.com/documento/dc_36479aca0b52

How to cite

Chance-Constrained Personalized LLM Routing with Online Preference Learning — arXiv:2610.22695 — arXiv (en), https://arxiv.org/abs/2610.22695 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_36479aca0b52 (Metraipolis archive dc_36479aca0b52, sha256 bc2bbf0cbe36c3a8adf5f6a92dd56507b8235576d1f989ed8dfb0a26b96b5980)
BibTeX
@misc{dc36479aca0b52,
  title = {Chance-Constrained Personalized LLM Routing with Online Preference Learning — arXiv:2610.22695},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.22695},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_36479aca0b52; sha256 bc2bbf0cbe36c3a8adf5f6a92dd56507b8235576d1f989ed8dfb0a26b96b5980},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22695. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We study personalized LLM routing, where a router selects among heterogeneous models with different costs and capabilities. The router must minimize cumulative inference cost while ensuring reliable realized response quality, even though user preferences are unknown and must be learned from feedback during deployment. We formulate this problem as a constrained contextual bandit and develop an algorithm that learns preferences while balancing cost and quality. Under a logistic preference model and a strict-feasibility condition, we establish sublinear expected cost regret relative to an optimal stationary randomized contextual oracle and a high-probability guarantee on realized cumulative satisfaction. The framework extends beyond LLM routing to broader online model-selection problems with reliability requirements.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Applications (stat.AP).
Pagina del lavoro: https://arxiv.org/abs/2610.22695 — 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 2026-10-19T08:42:40.234Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).