{
  "abstract": "We study adaptive routing of prompts to large language model (LLM) experts to maximize response quality in an online setting with limited feedback. We formulate it as a bandit problem with $K$ actions that represent experts and $d$ features that encode prompts, over a horizon of $T$ rounds. We propose algorithms that strategically select and observe rewards to minimize regret. In the full-information setting, we achieve a regret of $\\tilde{O}(d T / \\sqrt{m})$, while in the bandit setting we achieve $\\tilde{O}(d T \\sqrt{K / m})$, where $m \\ll T$ is a budget on feedback. Our experiments show that we efficiently learn high-quality routing strategies across diverse LLMs from limited feedback.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.05820",
  "author_total": 7,
  "authors": [
    {
      "name": "Wang Wei"
    },
    {
      "name": "Soumyabrata Pal"
    },
    {
      "name": "Koyel Mukherjee"
    },
    {
      "name": "Franck Dernoncourt"
    },
    {
      "name": "Ryan A. Rossi"
    },
    {
      "name": "Branislav Kveton"
    },
    {
      "name": "Hoda Eldardiry"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/f60x0drqvq2p5.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.05820",
  "fulltext_url": null,
  "id": "f60x0drqvq2p5",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.05820",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1803|Advanced Bandit Algorithms Research"
    ]
  ],
  "openalex": "W7212035784",
  "pdf_url": "https://arxiv.org/pdf/2609.05820",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-05",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/f60x0drqvq2p5.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:20:07Z",
      "native_id": "2609.05820",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.05820"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Online Learning with LLM Experts from Limited Feedback",
  "title_text": "Online Learning with LLM Experts from Limited Feedback",
  "type": "preprint",
  "updated_at": "2026-10-30T12:20:07Z",
  "url": "https://research.draconitic.com/paper/f60x0drqvq2p5",
  "venue": "arXiv",
  "versions": []
}