{
  "abstract": "Single-cell foundation models perform well on cell-type classification, but their value for disease-state prediction and their advantage over classical representations remains unclear. We benchmarked cell-representations from four foundation models using logistic regression, on 868 disease-prediction tasks (82 datasets) and 65 cell-type datasets from CELLxGENE. With full data, the baseline (log-normalized counts) outperformed the best model (TranscriptFormer-Sapiens) for cell type and matched it for disease. However, with few labeled cells, TranscriptFormer-Sapiens significantly outperformed the baseline in both tasks up to ~102 samples per class. Our comprehensive zero-shot benchmark identifies the data-scarce regime as a domain where Single-cell foundation models truly show advantage.",
  "abstract_shown": true,
  "affiliations": [
    "The Institute for Medical Research Israel-Canada, Department of Developmental Biology and Cancer Research, Jerusalem, Israel"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "H. Benhayon"
    },
    {
      "name": "M. M. Danziger"
    },
    {
      "name": "D. Ben-Zvi"
    },
    {
      "name": "M. Rosen-Zvi"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/5p78rz05cbf51.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.10.12.758421",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.10.12.758421v1.full",
  "id": "5p78rz05cbf51",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.10.12.758421v1",
  "language": null,
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "biorxiv",
      "bioinformatics"
    ],
    [
      "openalex",
      "1312|Single-cell and spatial transcriptomics"
    ]
  ],
  "openalex": "W7221555460",
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.10.12.758421v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-12",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/5p78rz05cbf51.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:20Z",
      "native_id": "10.64898/2026.10.12.758421",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.10.12.758421v1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Benchmarking single-cell foundation models reveals an advantage in the low-data regime for cell-type and disease-state prediction",
  "title_text": "Benchmarking single-cell foundation models reveals an advantage in the low-data regime for cell-type and disease-state prediction",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:20Z",
  "url": "https://research.draconitic.com/paper/5p78rz05cbf51",
  "venue": "bioRxiv",
  "versions": []
}