# BrainCodes: A Lightweight EEG Foundation Model over Discrete Neural Codes

DevFeed: [BrainCodes: A Lightweight EEG Foundation Model over Discrete Neural Codes](<https://devfeed.tech/articles/braincodes-a-lightweight-eeg-foundation-model-over-discrete-neural-codes-94909.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.29320>)

Author: Kleanthis Avramidis, Ivan Tashev, Dimitra Emmanouilidou

Published: 2026-10-22T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [quantization](<https://devfeed.tech/topics/quantization.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [attention](<https://devfeed.tech/tags/attention.md>), [cognitive-load](<https://devfeed.tech/tags/cognitive-load.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ne](<https://devfeed.tech/tags/cs-ne.md>), [eeg](<https://devfeed.tech/tags/eeg.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [self-supervised-learning](<https://devfeed.tech/tags/self-supervised-learning.md>), [sleep](<https://devfeed.tech/tags/sleep.md>), [wearable](<https://devfeed.tech/tags/wearable.md>)

## AI overview

BrainCodes is a compact EEG foundation model pretrained with self-supervised learning on more than 800 hours of recordings from over 1,000 participants. It combines a quantization-based neural tokenizer with a spatiotemporal encoder designed for arbitrary electrode configurations; its training encourages spatial connectivity features and sensitivity to narrowband oscillations. The abstract reports performance matching or exceeding the state of the art on sleep staging, attention assessment, and cognitive load estimation, with transfer to low-resource wearable devices using up to 97% fewer trainable parameters than comparable architectures. The excerpt also describes evaluation of learned representation geometry and invariances across spatial, temporal, and spectral signal properties.

## Source excerpt

arXiv:2610.29320v1 Announce Type: new Abstract: Recent electroencephalography (EEG) foundation models increasingly emphasize model scale, potentially limiting their deployment on resource-constrained devices and in everyday environments. We argue that transferable EEG features can instead be learned with a compact architecture and introduce BrainCodes, a lightweight foundation model pretrained through self-supervised learning on more than 800 hours of recordings from over 1,000 participants. Our architecture combines a quantization-based neural tokenizer with a spatiotemporal encoder that supports arbitrary electrode configurations. The pretraining objective explicitly encourages representations that capture spatial connectivity through channel positional encoding, and sensitivity to narrowband neural oscillations through an auxiliary training objective. Our model matches or surpasses the state of the art across diverse neurophysiological monitoring tasks, including sleep staging, attention assessment, and cognitive load estimation. BrainCodes transfers effectively to low-resource wearable devices while using up to 97% fewer trainable parameters than comparable architectures. Beyond predictive accuracy, our evaluation framework characterizes the geometry and invariances of learned EEG representations across measurable spatial, temporal, and spectral signal properties.