Pith. sign in

REVIEW 27 cited by

Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCI

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.18765 v1 pith:27N76RX4 submitted 2024-05-29 cs.LG

Wei-Bang Jiang , Li-Ming Zhao , Bao-Liang Lu This is my paper · ORCID
classification cs.LG
keywords modelsneuraldatasetslabramlargecapabilitieschanneldata
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

The current electroencephalogram (EEG) based deep learning models are typically designed for specific datasets and applications in brain-computer interaction (BCI), limiting the scale of the models and thus diminishing their perceptual capabilities and generalizability. Recently, Large Language Models (LLMs) have achieved unprecedented success in text processing, prompting us to explore the capabilities of Large EEG Models (LEMs). We hope that LEMs can break through the limitations of different task types of EEG datasets, and obtain universal perceptual capabilities of EEG signals through unsupervised pre-training. Then the models can be fine-tuned for different downstream tasks. However, compared to text data, the volume of EEG datasets is generally small and the format varies widely. For example, there can be mismatched numbers of electrodes, unequal length data samples, varied task designs, and low signal-to-noise ratio. To overcome these challenges, we propose a unified foundation model for EEG called Large Brain Model (LaBraM). LaBraM enables cross-dataset learning by segmenting the EEG signals into EEG channel patches. Vector-quantized neural spectrum prediction is used to train a semantically rich neural tokenizer that encodes continuous raw EEG channel patches into compact neural codes. We then pre-train neural Transformers by predicting the original neural codes for the masked EEG channel patches. The LaBraMs were pre-trained on about 2,500 hours of various types of EEG signals from around 20 datasets and validated on multiple different types of downstream tasks. Experiments on abnormal detection, event type classification, emotion recognition, and gait prediction show that our LaBraM outperforms all compared SOTA methods in their respective fields. Our code is available at https://github.com/935963004/LaBraM.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 27 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Pretraining Large Brain Language Model for Active BCI: Silent Speech

    cs.CL 2025-04 conditional novelty 7.0 of 10

    Autoregressive spectro-temporal pretraining of an EEG transformer improves silent speech decoding accuracy in cross-session tests, with a new 120-hour dataset.

  2. EEG-PRIME: Prototype-Aligned Representation Learning with Multi-Level Conditioning for EEG Decoding

    cs.AI 2026-08 conditional novelty 6.0 of 10

    EEG-PRIME aligns EEG signals with frozen text embeddings of class labels under task, dataset, and subject-invariance conditioning, achieving strong cross-dataset decoding and competitive zero-shot motor-imagery transfer.

  3. Joint Text-Audio Alignment for EEG-to-Text Decoding in Chinese Speech Production and Perception

    cs.AI 2026-07 conditional novelty 6.0 of 10

    Joint text-audio contrastive alignment plus CTC decoding yields state-of-the-art closed-set Chinese sentence identification from scalp EEG: 82.37% top-1 on reading-aloud and 41.43% on passive-listening EEG (101 candidates).

  4. Cross-Cohort Spectral-Temporal Dissociation in Frozen EEG Foundation-Model Representations

    q-bio.NC 2026-07 conditional novelty 6.0 of 10

    Frozen EEG foundation models can decode the aperiodic spectral slope, but not the alpha-envelope DFA exponent reproducibly across two cohorts.

  5. Physiological Noise Augmentation Improves Non-Invasive Brain-to-Speech

    cs.LG 2026-07 conditional novelty 6.0 of 10

    PNA decomposes MEG recordings via ICA, isolates artifact components using EOG/ECG references, and re-injects scaled artifacts into clean data to train decoders that are invariant to physiological noise, improving imag...

  6. Masked Generative-Contrastive Representation Learning for Cross-Dataset EEG-Based Emotion Recognition

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Pretraining a region-aware spatiotemporal EEG encoder with JEPA generative and masked dynamic contrastive losses on FACED yields higher cross-subject accuracy than SSL baselines when fine-tuned on SEED-IV/V/VII.

  7. Escaping the BLEU Trap: A Signal-Grounded Framework with Decoupled Semantic Guidance for EEG-to-Text Decoding

    cs.CL 2026-02 unverdicted novelty 6.0 of 10

    SemKey predicts four semantic attributes from EEG and conditions a frozen LLM on them, beating prior decoders on new semantic-alignment metrics while leaving true word-level accuracy low (2.7% content recall).

  8. Data Normalization Strategies for EEG Deep Learning

    eess.SP 2025-06 conditional novelty 6.0 of 10

    Window-level, per-channel normalization helps supervised EEG tasks, while minimal or cross-channel window normalization suits contrastive self-supervised learning on EEG.

  9. DIVER-0 : A Fully Channel Equivariant EEG Foundation Model

    eess.SP 2025-06 conditional novelty 6.0 of 10

    A channel-permutation-equivariant EEG transformer with full spatio-temporal attention achieves competitive BCI performance with only 10% of pretraining data.

  10. ZUNA1.1: A more flexible EEG foundation model for Denoising and Super-resolution

    cs.LG 2026-07 conditional novelty 5.0 of 10

    ZUNA1.1, an open-source 380M EEG diffusion autoencoder, reconstructs variable-length, flexibly masked EEG at least as well as its predecessor and far better than spherical spline interpolation.

  11. STST-JEPA: Shallow-Target Spatio-Temporal Joint Embedding Prediction Architecture For EEG Self-Supervised Learning

    cs.LG 2026-07 conditional novelty 5.0 of 10

    A JEPA-style EEG foundation model with shallow EMA targets plus light reconstruction reaches strong multi-task transfer and 3.06-year validation age MAE on a large multi-site corpus.

  12. BEAM: Brainwave Empathy Assessment Model for Early Childhood

    cs.LG 2025-09 conditional novelty 5.0 of 10

    BEAM, a multi-view EEG deep learning model, predicts high vs low empathy in 4-6 year olds with 64.7% accuracy and 0.008 standard deviation on 57 children.

  13. IMU-Enhanced EEG Motion Artifact Removal with Fine-Tuned Large Brain Models

    cs.LG 2025-09 reject novelty 5.0 of 10

    A LaBraM-based model with IMU attention mapping is fine-tuned for EEG motion artifact removal, but its evaluation metric is identical to its training loss and no independent check of neural signal preservation is provided.

  14. Transformer-based EEG Decoding: A Survey

    cs.LG 2025-07 conditional novelty 5.0 of 10

    A survey that classifies Transformer-based EEG decoding models into backbone, hybrid, and customized categories and reviews their applications and limitations.

  15. Cross-Modal Epileptic Signal Harmonization: Frequency Domain Mapping Quantization for Pre-training a Unified Neurophysiological Transformer

    q-bio.NC 2025-06 conditional novelty 5.0 of 10

    EpiNT, a transformer pretrained on over 2,700 hours of EEG and iEEG from 1,199 patients with masked autoencoding and a frequency-domain quantizer, matches or beats other pretrained models on six epilepsy classificatio...

  16. CRIA: A Cross-View Interaction and Instance-Adapted Pre-training Framework for Generalizable EEG Representations

    cs.LG 2025-06 conditional novelty 5.0 of 10

    Fusing temporal, spectral, and spatial EEG views with cross-attention and view-wise masking improves downstream classification and cross-dataset generalization over prior EEG pretraining models.

  17. From Theory to Application: Fine-Tuning Large EEG Model with Real-World Stress Data

    cs.LG 2025-05 conditional novelty 5.0 of 10

    Fine-tuning the LaBraM EEG foundation model on 82 real-world classroom stress recordings yields 90.47% balanced accuracy in the best of four seeds, but the tiny, seed-dependent test set makes the headline fragile.

  18. BrainStratify: Coarse-to-Fine Disentanglement of Intracranial Neural Dynamics

    eess.SP 2025-05 conditional novelty 5.0 of 10

    BrainStratify's coarse-to-fine disentanglement, electrode clustering plus decoupled product quantization, modestly improves speech decoding over prior methods on sEEG and epidural ECoG datasets.

  19. Differentiable Logic Gate Networks for Low-Latency EEG Classification on Edge Devices

    cs.LG 2026-07 conditional novelty 4.0 of 10

    Diff-Logic gate networks beat MLPs on dementia EEG classification and run nearly 3x faster and 14x smaller on edge hardware, though emotion-recognition gains are mixed.

  20. Evaluation of Stress Detection as Time Series Events -- A Novel Window-Based F1-Metric

    cs.LG 2025-09 conditional novelty 4.0 of 10

    A window-based F1 metric that rewards detections within a time tolerance reveals statistically significant stress-event prediction where pointwise F1 scores are all zero.

  21. Cross-BCI, A Cross-BCI-Paradigm Classifica-tion Model Towards Universal BCI Applications

    q-bio.QM 2025-08 reject novelty 4.0 of 10

    A single lightweight CNN classifies EEG from three BCI paradigms with 88.39% accuracy on OpenBMI, beating EEGNet, DeepConvNet, EEG-Inception, and EEGITNet.

  22. Large Language Models for EEG: A Comprehensive Survey and Taxonomy

    eess.SP 2025-06 conditional novelty 4.0 of 10

    A taxonomy and review of studies applying large language models to EEG signals, organized into four domains and three adaptation strategies.

  23. Real-World fNIRS-Based Brain-Computer Interfaces: Benchmarking Deep Learning and Classical Models in Interactive Gaming

    q-bio.NC 2025-05 reject novelty 4.0 of 10

    The paper reports high rest/task classification accuracy for fNIRS during a tennis game, but asymmetric augmentation of the rest class confounds the benchmark.

  24. Large Cognition Model: Towards Pretrained EEG Foundation Model

    eess.SP 2025-02 conditional novelty 4.0 of 10

    LCM, a transformer EEG model combining contrastive alignment and masked reconstruction, reports state-of-the-art balanced accuracy on BCIC-2A and BCIC-2B.

  25. AnyECG: Foundational Models for Multitask Cardiac Analysis in Real-World Settings

    eess.SP 2024-11 reject novelty 4.0 of 10

    AnyECG, a two-stage self-supervised ECG foundation model with a vector-quantized rhythm codebook and sparse attention, reports state-of-the-art numbers on four cardiac tasks, but its evaluation is weakened by in-domai...

  26. Technological Advances in Detecting and Managing Cognitive Impairment in Older Adults: Trends, Challenges, and Future Directions

    cs.LG 2026-07 conditional novelty 3.0 of 10

    Narrative review of cognitive-impairment detection technologies concludes that reported accuracies are often inflated by weak validation and that progress depends on multimodal, longitudinally validated, externally te...

  27. Large-Scale AI and Foundation Models for Neuroscience: A Comprehensive Review

    cs.AI 2025-10 conditional novelty 1.0 of 10

    This paper is a survey: it organizes existing foundation-model work in neuroscience into five application domains and lists public datasets, without presenting new experiments.

Pith tools