{
  "abstract": "Histology and transcriptomics provide complementary views of tissue biology, capturing spatial morphology and molecular activity, respectively. Pathology foundation models (PFMs) learn rich morphological representations from H&E images, yet strong downstream performance alone does not establish whether these representations encode biologically meaningful and robust molecular information. We present a **training-free framework for auditing biological alignment in frozen PFMs** using spatially paired histology and transcriptomics from HEST-1k, evaluated on **240 samples spanning three organs**. Multiple frozen PFMs are used to extract H&E representations, while gene expression is aggregated into biologically interpretable pathway-level programs. We use TabPFN as a pretrained probe to quantify the extent to which these molecular programs can be decoded from frozen image representations without task-specific gradient updates. Beyond predictive performance, our audit examines whether pathway decodability generalizes across tissue sections, patient groups, and tissue types; whether representations exhibit section-level or other shortcut dependencies; and whether predictions remain stable under small image perturbations and context resampling. This multi-tissue evaluation distinguishes molecular programs that are consistently encoded from those that are tissue-specific, unstable, or shortcut-sensitive. Our framework therefore provides a systematic approach for assessing not only **what biological information pathology foundation models encode, but also how reliably that information persists under clinically relevant sources of variation**.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.29523",
  "author_total": 6,
  "authors": [
    {
      "name": "Ushashi Bhattacharjee"
    },
    {
      "name": "Alloy Das"
    },
    {
      "name": "Saria Hannan"
    },
    {
      "name": "Tirtho Roy"
    },
    {
      "name": "Koushik Howlader"
    },
    {
      "name": "Soumik Sarkar"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/b8wr3q3eqn7xf.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.29523",
  "fulltext_url": null,
  "id": "b8wr3q3eqn7xf",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.29523",
  "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.MA"
    ],
    [
      "arxiv",
      "q-bio.QM"
    ],
    [
      "openalex",
      "1702|AI in cancer detection"
    ]
  ],
  "openalex": "W7214352256",
  "pdf_url": "https://arxiv.org/pdf/2609.29523",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-01",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/b8wr3q3eqn7xf.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:24:16Z",
      "native_id": "2609.29523",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.29523"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "arxiv:q-bio.QM",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Towards Trustworthy Biological Alignment in TabPFN-Probed Pathology Foundation Models",
  "title_text": "Towards Trustworthy Biological Alignment in TabPFN-Probed Pathology Foundation Models",
  "type": "preprint",
  "updated_at": "2026-10-30T12:24:16Z",
  "url": "https://research.draconitic.com/paper/b8wr3q3eqn7xf",
  "venue": "arXiv",
  "versions": []
}