Metraipolis › archive › dc_5de783f3391f

Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text — arXiv:2609.40359

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.40359
License (as archived)
CC0-1.0
Archived
2026-10-07T13:47:04.274Z
SHA-256 of the archived text
ccfe081b1266258000ea2f3ff95081b93eea0a98d27f638d53830f268ba96231
Stable URL of this page
https://metraipolis.com/documento/dc_5de783f3391f

How to cite

Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text — arXiv:2609.40359 — arXiv (en), https://arxiv.org/abs/2609.40359 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5de783f3391f (Metraipolis archive dc_5de783f3391f, sha256 ccfe081b1266258000ea2f3ff95081b93eea0a98d27f638d53830f268ba96231)
BibTeX
@misc{dc5de783f3391f,
  title = {Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text — arXiv:2609.40359},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.40359},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5de783f3391f; sha256 ccfe081b1266258000ea2f3ff95081b93eea0a98d27f638d53830f268ba96231},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40359. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We find that major reported improvements in decoding words from non-invasive brain recordings are largely reproducible without any brain data. In the influential work of d'Ascoli et al. (2025), time series of brain activity from subjects perceiving continuous speech are segmented into fixed-length windows starting at each word. A neural network then generates predictions for all of the words in a sentence together. Neighbouring windows partially overlap, implicitly revealing the interval between words. Since these intervals indicate the duration of the words spoken, and different words tend to have different durations - for example, "the" is much shorter than "supercalifragilisticexpialidocious" - the neural network can improve its predictions of words without relying on the underlying brain activity. Consistent with this, the method reaches 22.0% balanced accuracy on synthetic signals containing no brain information, compared with 22.3% on real brain recordings. To prevent the network from learning this shortcut, we make a single, simple change. Instead of jointly encoding all windows in a sentence, we process each independently. As a result, the neural network achieves better performance by learning underlying word-specific information from brain recordings. This makes two existing strategies become much more effective than before. Both aggregating predictions from distinct neural responses to the same word and using a pretrained LLM as a linguistic prior now substantially improve results. On our perceived speech benchmark, this simple recipe (SimpleB2T) achieves a word error rate of 36.6% with five observations per word, approaching past invasive speech decoding performance, albeit under different conditions. The results in this work expose an important shortcut in brain-to-text decoding and show that removing it leads to a simple and considerably more effective strategy.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Neurons and Cognition (q-bio.NC).
Pagina del lavoro: https://arxiv.org/abs/2609.40359 — 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 2026-10-07T13:18:10.534Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).