{
  "abstract": "Large-scale Neuropixels recordings sample many brain areas, but no single experimental session records every area of interest. NeuroPaint, a transformer-based autoencoder, integrates such recordings by aligning area-specific latent dynamics across sessions and \"inpainting\" activity in unrecorded areas. Could a linear model, which would be computationally efficient and potentially more interpretable, suffice to achieve this? To address this question, we introduce Linear NeuroPaint, a linear variant of NeuroPaint, and evaluate it on two multi-area Neuropixels datasets. Linear NeuroPaint achieves strong predictive performance on recorded areas, however, it is less accurate than nonlinear models at cross-area inpainting, despite extensive hyperparameter optimization. Moreover, its parameters are not readily interpretable: incomplete cross-session alignment leads parameter-defined neuron clusters to exhibit heterogeneous response profiles. Together, these results show that linear transformations are insufficient for cross-session alignment and cross-area prediction. Nevertheless, Linear NeuroPaint provides a computationally efficient baseline for benchmarking multi-session neural models.",
  "abstract_shown": true,
  "affiliations": [
    "Columbia University"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "name": "Y. J. Yao"
    },
    {
      "affs": [
        0
      ],
      "name": "J. Xia"
    },
    {
      "name": "L. Paninski"
    },
    {
      "name": "K. D. Miller"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/92efgeqar7z88.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.10.03.756082",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.10.03.756082v1.full",
  "id": "92efgeqar7z88",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.10.03.756082v1",
  "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",
      "neuroscience"
    ],
    [
      "openalex",
      "2805|Neural dynamics and brain function"
    ]
  ],
  "openalex": "W7221441456",
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.10.03.756082v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-03",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/92efgeqar7z88.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:04Z",
      "native_id": "10.64898/2026.10.03.756082",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.10.03.756082v2"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:neuroscience",
      "subfield": "neuroscience"
    }
  ],
  "title": "Linear NeuroPaint: An Efficient Baseline for Cross-Session Neural Alignment and Inpainting",
  "title_text": "Linear NeuroPaint: An Efficient Baseline for Cross-Session Neural Alignment and Inpainting",
  "type": "preprint",
  "updated_at": "2026-11-06T00:21:32Z",
  "url": "https://research.draconitic.com/paper/92efgeqar7z88",
  "venue": "bioRxiv",
  "versions": []
}