Metraipolis › archive › dc_0364efe056dc

The Visual Target Matters: Learning across the Visual Hierarchy for Brain-to-Image Retrieval — arXiv:2609.24136

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.24136
License (as archived)
CC0-1.0
Archived
2026-10-09T22:05:11.200Z
SHA-256 of the archived text
77fc8a742a21c2f4fd7018ab6903289f4384c0c8119d5c260789ef9f5361f9c9
Stable URL of this page
https://metraipolis.com/documento/dc_0364efe056dc

How to cite

The Visual Target Matters: Learning across the Visual Hierarchy for Brain-to-Image Retrieval — arXiv:2609.24136 — arXiv (en), https://arxiv.org/abs/2609.24136 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0364efe056dc (Metraipolis archive dc_0364efe056dc, sha256 77fc8a742a21c2f4fd7018ab6903289f4384c0c8119d5c260789ef9f5361f9c9)
BibTeX
@misc{dc0364efe056dc,
  title = {The Visual Target Matters: Learning across the Visual Hierarchy for Brain-to-Image Retrieval — arXiv:2609.24136},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.24136},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0364efe056dc; sha256 77fc8a742a21c2f4fd7018ab6903289f4384c0c8119d5c260789ef9f5361f9c9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.24136. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Brain-to-image retrieval seeks to identify the visual stimulus that elicited a non-invasive neural response. Candidate images are typically represented by pretrained vision models, whose internal representations vary in abstraction across depth. Existing methods usually train the neural encoder to recover a fixed final-layer visual target. Under this formulation, the visual hierarchy is reduced to a single prescribed endpoint, preventing representations at other depths from directly shaping the visual target. This limitation motivates learning how information across visual depths should contribute to the retrieval target. To this end, we introduce NeuroGlyph, which learns a trial-independent visual target from multiple depths of a frozen visual backbone. NeuroGlyph decomposes the target into factor-specific subspaces. Each subspace learns an image-conditioned allocation over visual depth. The resulting subspaces are fused into a single embedding for retrieval. Across THINGS-EEG and THINGS-MEG, NeuroGlyph outperforms final-layer supervision in all controlled comparisons. It also surpasses the post hoc best fixed-layer oracle in three of four comparisons. Parameter-matched ablations support both factorized target construction and image-conditioned depth allocation. Under comparable 200-way retrieval protocols, NeuroGlyph achieves the strongest system-level performance in six of eight reported metrics. These results support learning retrieval targets across the visual hierarchy rather than prescribing one visual depth.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.24136 — 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 2026-10-09T21:45:20.993Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).