{
  "abstract": "Laboratories often face a new molecular assay with 16-64 labels and a bank of predictors whose training data and parameters are unavailable. The practical question is which frozen outputs to include in a small local model. AssayRouter treats completed assays as pseudo-targets and labels each candidate by its post-fit utility: the reduction in held-out discovery loss when the candidate is added to the local target predictor. A shared regressor learns to predict this utility from candidate behavior on the support set, without source identity; on a new assay, one frozen ranking selects four sources and separate labels fit a convex combiner. We train only on completed ChEMBL-MT assays and evaluate 24 external regression assays across six frozen interface families. AssayRouter-C lowers strict four-call negative log-likelihood (NLL) by 0.0409 relative to Support-CV@4. Frozen candidate-label permutations confirm that candidate-utility correspondence carries the transferred information, and leave-one-interface-out training shows that the mapping generalizes to unseen predictor families. Completed assays therefore provide transferable supervision for scarce-label routing through frozen prediction interfaces.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.37285",
  "author_total": 8,
  "authors": [
    {
      "name": "Dong Xu"
    },
    {
      "name": "Zhangfan Yang"
    },
    {
      "name": "Jiantao Wu"
    },
    {
      "name": "Shipeng Zhang"
    },
    {
      "name": "Zexuan Zhu"
    },
    {
      "name": "Jiangqiang Li"
    },
    {
      "name": "Jun Zhang"
    },
    {
      "name": "Junkai Ji"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/967pq9jxzw8bq.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.37285",
  "fulltext_url": null,
  "id": "967pq9jxzw8bq",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.37285",
  "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.AI"
    ],
    [
      "arxiv",
      "q-bio.BM"
    ],
    [
      "arxiv",
      "q-bio.QM"
    ],
    [
      "openalex",
      "1703|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W7214968298",
  "pdf_url": "https://arxiv.org/pdf/2609.37285",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-29",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/967pq9jxzw8bq.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:25:36Z",
      "native_id": "2609.37285",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.37285"
    }
  ],
  "subjects": [
    {
      "field": "biochemistry",
      "origin": "arxiv:q-bio.BM",
      "subfield": "biophysics"
    },
    {
      "field": "biology",
      "origin": "arxiv:q-bio.QM",
      "subfield": "bioinformatics"
    }
  ],
  "title": "AssayRouter: Historical Utility Priors for Frozen Molecular Predictor Routing",
  "title_text": "AssayRouter: Historical Utility Priors for Frozen Molecular Predictor Routing",
  "type": "preprint",
  "updated_at": "2026-10-30T12:25:36Z",
  "url": "https://research.draconitic.com/paper/967pq9jxzw8bq",
  "venue": "arXiv",
  "versions": []
}