{"ID":2863556,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.24700","arxiv_id":"2509.24700","title":"A Robust Multi-Scale Framework with Test-Time Adaptation for sEEG-Based Speech Decoding","abstract":"Decoding speech from stereo-electroencephalography (sEEG) signals has emerged as a promising direction for brain-computer interfaces (BCIs). Its clinical applicability, however, is limited by the inherent non-stationarity of neural signals, which causes domain shifts between training and testing, undermining decoding reliability. To address this challenge, a two-stage framework is proposed for enhanced robustness. First, a multi-scale decomposable mixing (MDM) module is introduced to model the hierarchical temporal dynamics of speech production, learning stable multi-timescale representations from sEEG signals. Second, a source-free online test-time adaptation (TTA) method performs entropy minimization to adapt the model to distribution shifts during inference. Evaluations on the public DU-IN spoken word decoding benchmark show that the approach outperforms state-of-the-art models, particularly in challenging cases. This study demonstrates that combining invariant feature learning with online adaptation is a principled strategy for developing reliable BCI systems. Our code is available at https://github.com/lyyi599/MDM-TENT.","short_abstract":"Decoding speech from stereo-electroencephalography (sEEG) signals has emerged as a promising direction for brain-computer interfaces (BCIs). Its clinical applicability, however, is limited by the inherent non-stationarity of neural signals, which causes domain shifts between training and testing, undermining decoding r...","url_abs":"https://arxiv.org/abs/2509.24700","url_pdf":"https://arxiv.org/pdf/2509.24700v1","authors":"[\"Suli Wang\",\"Yang-yang Li\",\"Siqi Cai\",\"Haizhou Li\"]","published":"2025-09-29T12:31:57Z","proceeding":"cs.HC","tasks":"[\"cs.HC\"]","methods":"[]","has_code":false,"code_links":[{"ID":609021,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2863556,"paper_url":"https://arxiv.org/abs/2509.24700","paper_title":"A Robust Multi-Scale Framework with Test-Time Adaptation for sEEG-Based Speech Decoding","repo_url":"https://github.com/lyyi599/MDM-TENT","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}