| title | What masking geometry works best for EEG foundation models? |
|---|---|
| authors | Pierre Guetschel, Bruno Aristimunha, Yassine El Ouahidi, Arnaud Delorme, Thomas Moreau, Michael Tangermann |
| date | 2026-09-27 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2609.33487 |
| affiliation | Hugging Face |
| upvotes | 3 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | EEG foundation models hold promise for scalable brain-signal decoding across clinical and cognitive neuroscience applications, yet their pre-training pipelines remain poorly understood. Among design choices, the masking strategy is particularly critical: it determines what the network must predict and from which context. Yet it has never been ablated in isolation, as each new model bundles a new masking strategy with a new backbone and objective. In this paper, we formalize the design choices for spatio-temporal masking strategies and train various models with a single pipeline under varying masking configurations across two SSL frameworks (MAE and JEPA). We then systematically evaluate the resulting 58 pre-trained models on the 12 datasets of OpenEEGBench under a linear probe. Both frameworks agree on an optimal masking configuration and on shared failure modes. Outside these, performance is robust: 11 MAE and 9 JEPA configurations are statistically indistinguishable from the best. We further identify a novel JEPA-specific failure mode, tagged bias-inflation collapse, invisible to standard detectors. With a well-chosen mask, our pipeline reaches REVE-level downstream performance at a fraction of REVE's pre-training compute. |
cite
@misc{guetschel2026what,
title={What masking geometry works best for EEG foundation models?},
author={Pierre Guetschel and Bruno Aristimunha and Yassine El Ouahidi and Arnaud Delorme and Thomas Moreau and Michael Tangermann},
year={2026},
eprint={2609.33487},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2609.33487}
}