Metraipolis › archive › dc_4f33ee670cbb

Shared Is Not Enough: Predictor Width and Factor Readability Under Masked Latent Prediction — arXiv:2610.32562

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.32562
License (as archived)
CC0-1.0
Archived
2026-10-26T20:12:17.243Z
SHA-256 of the archived text
630572460c1b6eac6af8afa341909e5fb7e113b54d10216eb2c639c226651214
Stable URL of this page
https://metraipolis.com/documento/dc_4f33ee670cbb

How to cite

Shared Is Not Enough: Predictor Width and Factor Readability Under Masked Latent Prediction — arXiv:2610.32562 — arXiv (en), https://arxiv.org/abs/2610.32562 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4f33ee670cbb (Metraipolis archive dc_4f33ee670cbb, sha256 630572460c1b6eac6af8afa341909e5fb7e113b54d10216eb2c639c226651214)
BibTeX
@misc{dc4f33ee670cbb,
  title = {Shared Is Not Enough: Predictor Width and Factor Readability Under Masked Latent Prediction — arXiv:2610.32562},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.32562},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4f33ee670cbb; sha256 630572460c1b6eac6af8afa341909e5fb7e113b54d10216eb2c639c226651214},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32562. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV); Data Analysis, Statistics and Probability (physics.data-an).
Pagina del lavoro: https://arxiv.org/abs/2610.32562 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-26T19:49:03.413Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).