{"slug": "removing-timing-shortcuts-improves-non-invasive-brain-to-text", "title": "Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text", "summary": "A September 30, 2026 arXiv paper reports that major gains in non-invasive brain-to-text decoding are largely reproducible without any brain data, because fixed-length windows starting at each word implicitly leak word-duration timing cues. The authors show the method of d'Ascoli et al. (2025) reaches 22.0% balanced accuracy on synthetic signals containing no brain information versus 22.3% on real brain recordings, and that processing each window independently instead of jointly encoding a sentence removes the shortcut. Their SimpleB2T recipe achieves a 36.6% word error rate with five observations per word on a perceived-speech benchmark, approaching past invasive speech decoding performance under different conditions.", "body_md": "# Computer Science > Machine Learning\n\n  [Submitted on 30 Sep 2026]\n\n# Title:Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text\n\n[View PDF](https://arxiv.org/pdf/2609.40359)\n\n[HTML (experimental)](https://arxiv.org/html/2609.40359v1)\n\nAbstract: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.\n    \n\n### Current browse context:\n\ncs.LG\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\nIArxiv Recommender\n\n*(*[What is IArxiv?](https://iarxiv.org/about))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/removing-timing-shortcuts-improves-non-invasive-brain-to-text", "canonical_source": "https://arxiv.org/abs/2609.40359", "published_at": "2026-10-02 06:07:11+00:00", "updated_at": "2026-10-02 06:15:42.888032+00:00", "lang": "en", "topics": ["machine-learning", "ai-research", "natural-language-processing", "neural-networks"], "entities": ["d'Ascoli et al. (2025)", "SimpleB2T", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/removing-timing-shortcuts-improves-non-invasive-brain-to-text", "markdown": "https://wpnews.pro/news/removing-timing-shortcuts-improves-non-invasive-brain-to-text.md", "text": "https://wpnews.pro/news/removing-timing-shortcuts-improves-non-invasive-brain-to-text.txt", "jsonld": "https://wpnews.pro/news/removing-timing-shortcuts-improves-non-invasive-brain-to-text.jsonld"}}