New EEG Foundation Models Tackle Long Sequences and Montage Variability · 10 sources tracked
ByPulseAugur Editorial·[11 sources]·
Multiple research papers published on arXiv explore advancements in electroencephalography (EEG) foundation models, focusing on improving their ability to process long sequences and generalize across diverse datasets. NeurDuo-EEG introduces a model with persistent memory for continuous EEG analysis, achieving state-of-the-art results on several benchmarks. CortexBridge offers a lightweight adapter to align arbitrary EEG montages into a shared cortical latent space, enhancing performance for existing foundation models. NeuroAtlas presents a large-scale benchmark for clinical EEG and brain-computer interfaces, highlighting limitations in current models and the need for clinically relevant metrics. Other research investigates representational robustness to electrode montages, graph-structured rectified flows for dynamic modeling, physiology-guided hierarchical models, and methods to obstruct shortcut learning in EEG foundation models.
AI
IMPACT
Advances in EEG foundation models could lead to more accurate and interpretable diagnostics for neurological disorders and improved brain-computer interfaces.
RANK_REASON
Multiple research papers published on arXiv detailing new models and benchmarks for EEG analysis.
arXiv:2609.38587v1 Announce Type: new Abstract: Electroencephalography (EEG) is recorded continuously over hours, with relevant dynamics spanning timescales from milliseconds to hours. Most EEG foundation models nevertheless process fixed windows independently, limiting their abi…
arXiv:2610.01124v1 Announce Type: new Abstract: Electroencephalography (EEG) foundation models are often pretrained with a fixed channel vocabulary or a limited set of montages, making transfer difficult when electrode layouts change. We propose CortexBridge, a lightweight adapte…
arXiv cs.AI
TIER_1English(EN)·Konstantinos Kontras, Trui Osselaer, Stylianos G. Mouslech, Angeliki-Ilektra Karaiskou, Guido Gagliardi, Thomas Strypsteen, Mohammad Hossein Badiei, Anku Rani, Maarten Vanmarcke, Miguel Bhagubai, Chanakya Ekbote, Jaedong Hwang, Christos Chatzichristos, P…·
arXiv:2605.14698v2 Announce Type: replace-cross Abstract: Foundation models (FMs) promise to extract unified representations that generalize across downstream tasks. They have emerged across fields, including electroencephalography (EEG), but it is less clear how effective they a…
arXiv cs.LG
TIER_1English(EN)·Jakob Steglich, Justus Meyer zu Bexten, Shakiba Moradi, Laure Ciernik, Simon M. Hofmann, Mina Jamshidi Idaji·
arXiv:2609.36288v1 Announce Type: new Abstract: EEG foundation models (EEG-FMs) are intended to generalize across different datasets by learning representations that, ideally, are invariant to dataset-specific EEG configurations such as electrode montages. However, EEG-FMs that a…
arXiv:2609.37934v1 Announce Type: new Abstract: Forecasting time-varying functional connectivity from electroencephalography (EEG) requires modeling both history-dependent trends and structured variability across channels. Conditional flow matching provides a framework for distri…
arXiv cs.AI
TIER_1English(EN)·Yipeng Zhang, Chenda Duan, Yuanyi Ding, Tianyi Wang, Atsuro Daida, Masaki Izumi, Yuta Tanoue, Naoto Kuroda, Shaun A. Hussain, Nishant Sinha, Eishi Asano. Hiroki Nariai, Vwani Roychowdhury·
arXiv:2609.36087v1 Announce Type: cross Abstract: Clinicians and neuroscientists have long analyzed intracranial electroencephalography (iEEG) through directly measurable physiological characteristics, which carry much of the information that downstream tasks depend on. Recent iE…
Forecasting time-varying functional connectivity from electroencephalography (EEG) requires modeling both history-dependent trends and structured variability across channels. Conditional flow matching provides a framework for distributional forecasting, yet it remains unclear whe…
Clinical scalp electroencephalography (EEG) offers a noninvasive window into neural dynamics of neuropsychiatric disorders. However, discriminative deep models often lack anatomically indexed physiological interpretability. We propose NeuroDyn-EEG, a pretraining framework integra…
arXiv:2609.31167v1 Announce Type: new Abstract: EEG foundation models increasingly use masked prediction to learn from unlabeled recordings, but optimizing this objective does not ensure transferable neural representations. A central challenge is that stable positional cues and l…
arXiv:2609.36971v1 Announce Type: new Abstract: Cross-subject EEG-to-image retrieval requires a neural represen- tation trained on source subjects to remain aligned with a visual embedding space for an unseen subject. Whereas existing methods primarily focus on the EEG side, we a…