{
  "abstract": "Vision encoders for Cell Painting are typically ranked by a single evaluation, commonly replicate mean average precision (mAP). We introduce CP-BG-Bench, a paired-view evaluation framework that holds the central cell fixed across four matched views (raw crop C, segmented S, and density-augmented variants CD and SD), ablating or augmenting surrounding pixels as a controlled intervention. Instantiating the framework on three datasets (JUMP-CP, RxRx1, RxRx3-core) and three encoders (DINOv3 ViT-B/16, OpenPhenom, SubCell) under four community-standard protocols (replicate mAP, scIB batch integration, CellProfiler feature prediction, cross-batch perturbation recall), we find that the four protocols rank the same encoders systematically differently, with disagreements decomposing along three axes: cell versus background, morphology versus context, and within-study versus across-batch. The largest effect: on RxRx3-core, SubCell with segmented inputs retains 94% of crop replicate mAP but only 32% of crop R@10, so the within-study signal preserved under segmentation is largely non-transferable; density augmentation recovers 84% of the within-study C-to-S gap but only 8% of the cross-batch gap. Segmented views predict CellProfiler features as well as or better than crops on two of three datasets, inverting the replicate-mAP ranking, and the C-to-S gap varies by an order of magnitude across datasets while remaining similar across encoders, indicating that background-driven gain is set by experimental design rather than by the encoder. Single-metric ranking of Cell Painting encoders is therefore sensitive to the protocol used, and protocol disagreements are interpretable as projections onto the three axes the paired-view design exposes. We will release the paired-view datasets, reconstruction pipelines, 36 trained checkpoints, aggregated embeddings, and the full evaluation suite.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.12761",
  "author_total": 5,
  "authors": [
    {
      "name": "Tim Treis"
    },
    {
      "name": "Nikita Moshkov"
    },
    {
      "name": "Johan Fredin Haslum"
    },
    {
      "name": "Shantanu Singh"
    },
    {
      "name": "Fabian J. Theis"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/cenqcnssmbmgf.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.12761",
  "fulltext_url": null,
  "id": "cenqcnssmbmgf",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.12761",
  "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.CV"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "q-bio.QM"
    ],
    [
      "openalex",
      "1304|Cell Image Analysis Techniques"
    ]
  ],
  "openalex": "W7213039722",
  "pdf_url": "https://arxiv.org/pdf/2609.12761",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-11",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/cenqcnssmbmgf.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:21:12Z",
      "native_id": "2609.12761",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.12761"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "arxiv:q-bio.QM",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Same Encoder, Different Winner: A Paired-View Framework for Cell Painting Encoder Evaluation",
  "title_text": "Same Encoder, Different Winner: A Paired-View Framework for Cell Painting Encoder Evaluation",
  "type": "preprint",
  "updated_at": "2026-10-30T12:21:12Z",
  "url": "https://research.draconitic.com/paper/cenqcnssmbmgf",
  "venue": "arXiv",
  "versions": []
}