{"slug": "spera-spherical-prior-eeg-foundation-model-with-geometry-and-frequency-aware", "title": "SPERA: Spherical Prior EEG Foundation Model with Geometry- and Frequency-Aware Latent Prediction", "summary": "SPERA, an EEG foundation model built on a joint-embedding predictive architecture (JEPA) that predicts in latent space, achieved the highest average balanced accuracy across nine downstream clinical, cognitive, and BCI tasks after pretraining on approximately 80,000 hours of EEG from 29,048 subjects across 106 datasets, according to the arXiv paper 2610.10571v1. SPERA adds a Legendre-polynomial spatial prior in attention to encode scalp electrode geometry, factorized temporal and spatial attention interleaved with periodic full-attention blocks, and a relational spectral regularizer aligning latent similarity with spectral views. The authors report strong parameter efficiency under linear probing and robustness across varying recording conditions, suggesting SPERA's potential as a general-purpose EEG backbone.", "body_md": "arXiv:2610.10571v1 Announce Type: new \nAbstract: Electroencephalography (EEG) provides a non-invasive measure of ongoing neural activity, but building general-purpose EEG models remains challenging due to the heterogeneity of subjects, devices, and electrode montages. Existing EEG foundation models predominantly rely on reconstruction-based objectives defined on the observed signal, which contains both neural and non-neural components. We introduce SPERA (Spherical Prior EEG Representation Architecture), an EEG foundation model that adopts the joint-embedding predictive architecture (JEPA) to predict in latent space. SPERA introduces a Legendre-polynomial spatial prior, incorporated into attention to encode varying scalp electrode geometries. Two further components adapt the model to EEG: factorized temporal and spatial attention interleaved with periodic full-attention blocks, and a relational spectral regularizer aligning latent similarity structure with spectral views. Pretrained on approximately 80,000 hours of EEG from 29,048 subjects across 106 datasets, SPERA achieves the highest average balanced accuracy across nine downstream tasks spanning clinical, cognitive, and BCI applications. SPERA further exhibits strong parameter efficiency under linear probing and robustness across varying recording conditions, suggesting its potential as a general-purpose backbone for diverse EEG analyses.", "url": "https://wpnews.pro/news/spera-spherical-prior-eeg-foundation-model-with-geometry-and-frequency-aware", "canonical_source": "https://arxiv.org/abs/2610.10571", "published_at": "2026-10-09 04:00:00+00:00", "updated_at": "2026-10-09 04:18:53.243235+00:00", "lang": "en", "topics": ["machine-learning", "ai-research", "neural-networks"], "entities": ["SPERA", "arXiv", "Legendre-polynomial spatial prior", "JEPA"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/spera-spherical-prior-eeg-foundation-model-with-geometry-and-frequency-aware", "markdown": "https://wpnews.pro/news/spera-spherical-prior-eeg-foundation-model-with-geometry-and-frequency-aware.md", "text": "https://wpnews.pro/news/spera-spherical-prior-eeg-foundation-model-with-geometry-and-frequency-aware.txt", "jsonld": "https://wpnews.pro/news/spera-spherical-prior-eeg-foundation-model-with-geometry-and-frequency-aware.jsonld"}}