{
  "abstract": "Off-policy evaluation (OPE) is essential for assessing decision policies before deployment when online experimentation is costly or unsafe. Accurate estimation remains difficult when behavior-policy data are limited and the target policy favors states and actions rarely observed in those data. We propose InCOPE, a prior-data fitted network that learns a shared policy-value prediction rule across synthetic OPE tasks. We construct an OPE task prior over Markov decision processes and behavior-target policy pairs, generating behavior-policy datasets and exact target-policy values for pretraining. For a new tabular decision problem, the pretrained network uses the behavior-policy dataset, target policy, and known task settings as context to predict the policy's expected return without task-specific parameter updates. Experiments across four tabular domains, including a semi-synthetic clinical setting, demonstrate InCOPE's effectiveness for policy-value estimation.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.34035",
  "author_total": 3,
  "authors": [
    {
      "name": "Shuguang Yu"
    },
    {
      "name": "Ting Li"
    },
    {
      "name": "Fan Zhou"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/aqpvek20nn97d.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.34035",
  "fulltext_url": null,
  "id": "aqpvek20nn97d",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.34035",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Reinforcement Learning in Robotics"
    ]
  ],
  "openalex": "W7223845342",
  "pdf_url": "https://arxiv.org/pdf/2610.34035",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-25",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/aqpvek20nn97d.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:35:49Z",
      "native_id": "2610.34035",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.34035"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Prior-Data Fitted Networks for In-Context Off-Policy Evaluation",
  "title_text": "Prior-Data Fitted Networks for In-Context Off-Policy Evaluation",
  "type": "preprint",
  "updated_at": "2026-10-30T12:35:49Z",
  "url": "https://research.draconitic.com/paper/aqpvek20nn97d",
  "venue": "arXiv",
  "versions": []
}