MSBraM: A Multi-scale Self-supervised Brain Foundation Model for Hierarchical EEG Dynamics Learning

• URL: http://arxiv.org/abs/2607.21402v1
• Date: Thu, 23 Jul 2026 15:03:04 GMT
• Title: MSBraM: A Multi-scale Self-supervised Brain Foundation Model for Hierarchical EEG Dynamics Learning
• Abstract summary: Self-supervised foundation models have recently shown strong potential for electroencephalogram (EEG)-based analysis.<n>Existing approaches struggle to capture the inherently multi-scale temporal structure of EEG signals.<n>We propose MSBraM, a Multi-Scale self-supervised Brain foundation Model designed to learn hierarchical EEG representations.
• Score: 16.184230178980705
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Self-supervised foundation models have recently shown strong potential for electroencephalogram (EEG)-based analysis. However, existing approaches struggle to capture the inherently multi-scale temporal structure of EEG signals, where local neural patterns and long-range dependencies jointly encode task-relevant information. This limitation hampers cross-scale representation learning and generalization across diverse downstream tasks. To address this challenge, we propose MSBraM, a Multi-Scale self-supervised Brain foundation Model designed to learn hierarchical EEG representations. MSBraM follows a two-stage pretraining framework. First, a multi-scale neural tokenizer discretizes raw EEG signals into semantic codes at different temporal resolutions via vector-quantized reconstruction. Second, the model is pretrained to predict masked codes using a curriculum multi-scale masking strategy, progressively integrating fine-grained local patterns with global temporal context. We pretrain MSBraM on over 2,400 hours of EEG data and evaluate it across 10 downstream tasks on 12 public datasets. Extensive experiments show that MSBraM achieves superior performance on other state-of-the-art pretrained models, demonstrating strong generalization and transferability. These results indicate that explicitly modeling multi-scale temporal dynamics is critical for effective EEG foundation models.

Related papers

• Toward a Multi-View Brain Network Foundation Model: Cross-View Consistency Learning Across Arbitrary Atlases [62.33465338932216]
  MV-BrainFM is a multi-view brain network foundation model designed to learn generalizable and scalable representations from brain networks constructed with arbitrary atlases.<n>Experiments on more than 20K subjects from 17 fMRI datasets show that MV-BrainFM consistently outperforms 14 existing brain network foundation models.
  arXiv  Detail & Related papers  (2026-03-20T11:55:00Z)
• BrainRVQ: A High-Fidelity EEG Foundation Model via Dual-Domain Residual Quantization and Hierarchical Autoregression [26.114257185901838]
  We propose BrainRVQ, a general-purpose EEG foundation model pre-trained on a large-scale corpus of clinical EEG data.<n>BrainRVQ features a Dual-Domain Residual Vector Quantization (DD-RVQ) tokenizer that disentangles temporal waveforms and spectral patterns into hierarchical discrete codes.
  arXiv  Detail & Related papers  (2026-02-18T23:30:36Z)
• THD-BAR: Topology Hierarchical Derived Brain Autoregressive Modeling for EEG Generic Representations [3.253716156877394]
  We propose a novel Topology Hierarchical Derived Brain Autoregressive Modeling (THD-BAR) for EEG generic representations.<n>The core innovation of THD-BAR lies in the introduction of the Brain Topology Hierarchy (BTH), which establishes a multi-scale spatial order for EEG channels.<n>Based on BTH, we design a Topology-Hierarchical Vector Quantized-Variational Autoencoder (THVQ-VAE) for multi-scale tokenization and develop an enhanced Brain Autoregressive (BAR) module with specialized masking strategies for prediction.
  arXiv  Detail & Related papers  (2025-11-05T13:20:14Z)
• NeuroRVQ: Multi-Scale EEG Tokenization for Generative Large Brainwave Models [66.91449452840318]
  We introduce NeuroRVQ, a scalable Large Brainwave Model (LBM) centered on a codebook-based tokenizer.<n>Our tokenizer integrates: (i) multi-scale feature extraction modules that capture the full frequency neural spectrum; (ii) hierarchical residual vector quantization (RVQ) codebooks for high-resolution encoding; and, (iii) an EEG signal phase- and amplitude-aware loss function for efficient training.<n>Our empirical results demonstrate that NeuroRVQ achieves lower reconstruction error and outperforms existing LBMs on a variety of downstream tasks.
  arXiv  Detail & Related papers  (2025-10-15T01:26:52Z)
• EEGDM: EEG Representation Learning via Generative Diffusion Model [17.595769291603688]
  We propose an EEG representation learning framework building upon Generative Diffusion Model (EEGDM)<n>Specifically, we developed a structured state-space model for diffusion pretraining and trained it using Denoising Diffusion Probabilistic Model (DDPM) framework.<n>The resulting latent EEG representations were then used for downstream classification tasks via our proposed latent fusion transformer (LFT)
  arXiv  Detail & Related papers  (2025-08-13T14:40:52Z)
• CSBrain: A Cross-scale Spatiotemporal Brain Foundation Model for EEG Decoding [57.90382885533593]
  We propose a Cross-scale Spatiotemporal Brain foundation model for generalized decoding EEG signals.<n>We show that CSBrain consistently outperforms task-specific and foundation model baselines.<n>These results establish cross-scale modeling as a key inductive bias and position CSBrain as a robust backbone for future brain-AI research.
  arXiv  Detail & Related papers  (2025-06-29T03:29:34Z)
• CodeBrain: Towards Decoupled Interpretability and Multi-Scale Architecture for EEG Foundation Model [52.466542039411515]
  EEG foundation models (EFMs) have emerged to address the scalability issues of task-specific models.<n>We present CodeBrain, a two-stage EFM designed to fill this gap.<n>In the first stage, we introduce the TFDual-Tokenizer, which decouples heterogeneous temporal and frequency EEG signals into discrete tokens.<n>In the second stage, we propose the multi-scale EEGSSM architecture, which combines structured global convolution with sliding window attention.
  arXiv  Detail & Related papers  (2025-06-10T17:20:39Z)
• BrainOmni: A Brain Foundation Model for Unified EEG and MEG Signals [46.121056431476156]
  This paper proposes Brain Omni, the first brain foundation model that generalises across heterogeneous EEG and MEG recordings.<n>Existing approaches typically rely on separate, modality- and dataset-specific models, which limits performance and cross-domain scalability.<n>A total of 1,997 hours of EEG and 656 hours of MEG data are curated and standardised from publicly available sources for pretraining.
  arXiv  Detail & Related papers  (2025-05-18T14:07:14Z)
• BrainGPT: Unleashing the Potential of EEG Generalist Foundation Model by Autoregressive Pre-training [15.135177893151008]
  EEGPT is the first generalist EEG foundation model designed to address these challenges.<n>First, we propose an electrode-wise modeling strategy that treats each electrode as a fundamental unit.<n>Second, we develop the first autoregressive EEG pre-trained model.<n>Third, we introduce a multi-task transfer learning paradigm using a learnable electrode graph network.
  arXiv  Detail & Related papers  (2024-10-14T12:17:54Z)
• PMT: Progressive Mean Teacher via Exploring Temporal Consistency for Semi-Supervised Medical Image Segmentation [51.509573838103854]
  We propose a semi-supervised learning framework, termed Progressive Mean Teachers (PMT), for medical image segmentation. Our PMT generates high-fidelity pseudo labels by learning robust and diverse features in the training process. Experimental results on two datasets with different modalities, i.e., CT and MRI, demonstrate that our method outperforms the state-of-the-art medical image segmentation approaches.
  arXiv  Detail & Related papers  (2024-09-08T15:02:25Z)
• MMA-DFER: MultiModal Adaptation of unimodal models for Dynamic Facial Expression Recognition in-the-wild [81.32127423981426]
  Multimodal emotion recognition based on audio and video data is important for real-world applications. Recent methods have focused on exploiting advances of self-supervised learning (SSL) for pre-training of strong multimodal encoders. We propose a different perspective on the problem and investigate the advancement of multimodal DFER performance by adapting SSL-pre-trained disjoint unimodal encoders.
  arXiv  Detail & Related papers  (2024-04-13T13:39:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.