Metraipolis › archive › dc_3f37063cdb68
Latent Prediction Yields Cross-Subject Alignable Representations in Sensory-Evoked fMRI — arXiv:2610.40280
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.40280
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T06:14:49.631Z
- SHA-256 of the archived text
6bc8d451e45aa8bb9c6f1a70771fad1397734bb5153ad58e176926ddae020f54- Stable URL of this page
https://metraipolis.com/documento/dc_3f37063cdb68
How to cite
Latent Prediction Yields Cross-Subject Alignable Representations in Sensory-Evoked fMRI — arXiv:2610.40280 — arXiv (en), https://arxiv.org/abs/2610.40280 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3f37063cdb68 (Metraipolis archive dc_3f37063cdb68, sha256 6bc8d451e45aa8bb9c6f1a70771fad1397734bb5153ad58e176926ddae020f54)
BibTeX
@misc{dc3f37063cdb68,
title = {Latent Prediction Yields Cross-Subject Alignable Representations in Sensory-Evoked fMRI — arXiv:2610.40280},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.40280},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3f37063cdb68; sha256 6bc8d451e45aa8bb9c6f1a70771fad1397734bb5153ad58e176926ddae020f54},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40280. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Decoding sensory-evoked functional magnetic resonance imaging (fMRI) data is a central analytical tool in cognitive neuroscience. Here we ask whether self-supervised learning (SSL), which has been scarcely explored in this domain, can yield representations that support strong downstream performance. To do so, we adapt a joint embedding predictive architecture (JEPA) to train participant-specific encoders on brain data alone. The encoders never see the stimulus images or any representation derived from them. Applied to the Natural Scenes Dataset (NSD), we evaluate our learned brain embeddings on the downstream tasks of brain captioning and image reconstruction, as well as cross-participant generalization. Our self-supervised approach achieves competitive performance with parameter-rich supervised pipelines that are trained end-to-end with access to image and language data. We further find that encoders trained independently for different participants can be aligned fully unsupervised, using neither images nor stimuli shared between them. After alignment, a neural decoding head trained on one participant can be applied to another without fine-tuning and retains most of its within-subject performance on image reconstruction and brain captioning. By removing the need for images and shared stimuli, our approach is a step towards pooling fMRI data across participants and, eventually, across datasets. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Neurons and Cognition (q-bio.NC). Pagina del lavoro: https://arxiv.org/abs/2610.40280 — 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 2026-10-31T06:07:08.761Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).