{
  "abstract": "Masked latent prediction methods such as I-JEPA aim to learn predictable features, but what determines which information they retain is less clear. We built a simulator in which every factor behind an image is known, so a probe can measure how readable each factor remains after training. We trained small encoders (0.7M parameters) with I-JEPA, SimCLR and BYOL on the same pairs of views and probed a factor that both views share. I-JEPA had a predictor with half the encoder's width and depth, the proportions it uses for ViT-B. Its median score was about 40% of SimCLR's. Widening the predictor to the encoder's width reversed this ordering at every training seed, against both other methods. We release the simulator and the trained models.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.32562",
  "author_total": 1,
  "authors": [
    {
      "name": "Nilotpal Kakati"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/bwqq96h9evspk.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.32562",
  "fulltext_url": null,
  "id": "bwqq96h9evspk",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.32562",
  "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",
      "physics.data-an"
    ],
    [
      "openalex",
      "1702|Domain Adaptation and Few-Shot Learning"
    ]
  ],
  "openalex": "W7223688839",
  "pdf_url": "https://arxiv.org/pdf/2610.32562",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-23",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/bwqq96h9evspk.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:35:15Z",
      "native_id": "2610.32562",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.32562"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "arxiv:physics.data-an",
      "subfield": "computational"
    }
  ],
  "title": "Shared Is Not Enough: Predictor Width and Factor Readability Under Masked Latent Prediction",
  "title_text": "Shared Is Not Enough: Predictor Width and Factor Readability Under Masked Latent Prediction",
  "type": "preprint",
  "updated_at": "2026-10-30T12:35:15Z",
  "url": "https://research.draconitic.com/paper/bwqq96h9evspk",
  "venue": "arXiv",
  "versions": []
}