LUNA: Efficient and Topology-Agnostic Foundation Model for EEG Signal Analysis

• URL: http://arxiv.org/abs/2510.22257v1
• Date: Sat, 25 Oct 2025 11:31:27 GMT
• Title: LUNA: Efficient and Topology-Agnostic Foundation Model for EEG Signal Analysis
• Abstract summary: We introduce LUNA, a self-supervised foundation model that reconciles disparate electrode geometries.<n>Luna compresses multi-channel EEG into a fixed-size, topology-agnostic latent space.<n>Luna transfers effectively to four downstream tasks: abnormality detection, artifact rejection, slowing classification, and emotion recognition.
• Score: 16.431555478831143
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Electroencephalography (EEG) offers a non-invasive lens into human brain activity, but building large-scale models is hampered by topological heterogeneity: each public EEG data defines its own electrode layout, limiting generalization. We introduce LUNA (Latent Unified Network Architecture), a self-supervised foundation model that reconciles disparate electrode geometries while scaling linearly -- not quadratically -- with channel count. LUNA compresses multi-channel EEG into a fixed-size, topology-agnostic latent space via learned queries and cross-attention. Downstream transformer blocks then operate exclusively on this latent representation using patch-wise temporal self-attention, decoupling computation from electrode count. Pre-trained on TUEG and Siena (over 21,000 hours of raw EEG across diverse montages) using a masked-patch reconstruction objective, LUNA transfers effectively to four downstream tasks: abnormality detection, artifact rejection, slowing classification, and emotion recognition. It demonstrates highly competitive performance across several benchmarks, achieving state-of-the-art results on TUAR and TUSL, e.g., 0.921 AUROC on TUAR, while reducing FLOPs by 300x and trimming GPU memory use by up to 10x. Critically, these gains are consistent across all evaluated electrode configurations. Code is available at https://github.com/pulp-bio/BioFoundation

Related papers

• Machine Learning-Based Reconstruction for Resistive Silicon Sensors [53.365453214937496]
  We study machine-learning-based reconstruction and compression for resistive silicon sensors.<n>We use full-waveform information from correlated pads to regularise the reconstruction and extract spatial information.<n>We first introduce recurrent neural network models based on LSTM layers, which provide a proof-of-concept implementation for full-waveform reconstruction.
  arXiv  Detail & Related papers  (2026-07-13T14:03:26Z)
• Towards Robust EEG Decoding Based on Riemannian Self-Attention [49.95065559566893]
  Brain-Computer Interface (BCI) based on electroencephalography (EEG) has significant applications in assistive technologies, medical rehabilitation, and entertainment.<n>Recent EEG decoding methods based on Symmetric Positive Definite (SPD) learning have demonstrated superior performance.<n>These methods typically employ basic network architectures and do not explicitly capture local relationships between EEG signals.<n>This limitation is problematic for EEG signals due to their inherently low Signal-to-Noise Ratio (SNR)
  arXiv  Detail & Related papers  (2026-06-24T06:31:20Z)
• B[FM]$^2$: Brain Foundation Model via Flow Matching with SplitUNet [33.19680160939894]
  Multi-channel EEG signals pose an architectural challenge for flow matching.<n>SplitUNet factorizes each block into separate 1D temporal and 1D electrode convolutions and downsamples only along time.<n>B[FM]$2$ sets a new state of the art on 7 of 9 standard downstream EEG classification tasks.
  arXiv  Detail & Related papers  (2026-06-18T18:01:26Z)
• LuMamba: Latent Unified Mamba for Electrode Topology-Invariant and Efficient EEG Modeling [14.663411219312827]
  We propose textbfLuMamba (textbfLatent textbfUnified textbfMamba), a self-supervised framework combining topology-invariant encodings with linear-complexity state-space modeling.<n>With only 4.6M parameters, LuMamba attains 80.99% balanced accuracy on TUAB and achieves state-of-art performance on Alzheimer's detection.
  arXiv  Detail & Related papers  (2026-03-19T16:26:11Z)
• REVE: A Foundation Model for EEG -- Adapting to Any Setup with Large-Scale Pretraining on 25,000 Subjects [5.368295573908306]
  REVE (Representation for EEG with Versatile Embeddings) is a pretrained model explicitly designed to generalize across diverse EEG signals.<n>We pretrain REVE on over 60,000 hours of EEG data from 92 datasets spanning 25,000 subjects, representing the largest EEG pretraining effort to date.<n>We release code, pretrained weights, and tutorials to support standardized EEG research and accelerate progress in clinical neuroscience.
  arXiv  Detail & Related papers  (2025-10-24T15:52:46Z)
• Multi-Channel Differential Transformer for Cross-Domain Sleep Stage Classification with Heterogeneous EEG and EOG [1.422074091840573]
  We propose SleepDIFFormer, a multi-channel differential transformer framework for heterogeneous EEG-EOG representation learning.<n>SleepDIFFormer is trained across multiple sleep staging datasets, each treated as a source domain, with the goal of generalizing to unseen target domains.<n>Our approach mitigates spatial and temporal attention noise and learns a domain-invariant EEG-EOG representation through feature distribution alignment across datasets, thereby enhancing generalization to new domains.
  arXiv  Detail & Related papers  (2025-08-21T03:59:10Z)
• Neural-Driven Image Editing [51.11173675034121]
  Traditional image editing relies on manual prompting, making it labor-intensive and inaccessible to individuals with limited motor control or language abilities.<n>We propose LoongX, a hands-free image editing approach driven by neurophysiological signals.<n>LoongX utilizes state-of-the-art diffusion models trained on a comprehensive dataset of 23,928 image editing pairs.
  arXiv  Detail & Related papers  (2025-07-07T18:31:50Z)
• 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)
• CEReBrO: Compact Encoder for Representations of Brain Oscillations Using Efficient Alternating Attention [46.47343031985037]
  We introduce a Compact for Representations of Brain Oscillations using alternating attention (CEReBrO)<n>Our tokenization scheme represents EEG signals at a per-channel patch.<n>We propose an alternating attention mechanism that jointly models intra-channel temporal dynamics and inter-channel spatial correlations, achieving 2x speed improvement with 6x less memory required compared to standard self-attention.
  arXiv  Detail & Related papers  (2025-01-18T21:44:38Z)
• Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCI [6.926908480247951]
  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. We then pre-train neural Transformers by predicting the original neural codes for the masked EEG channel patches.
  arXiv  Detail & Related papers  (2024-05-29T05:08:16Z)
• Physics-informed and Unsupervised Riemannian Domain Adaptation for Machine Learning on Heterogeneous EEG Datasets [53.367212596352324]
  We propose an unsupervised approach leveraging EEG signal physics. We map EEG channels to fixed positions using field, source-free domain adaptation. Our method demonstrates robust performance in brain-computer interface (BCI) tasks and potential biomarker applications.
  arXiv  Detail & Related papers  (2024-03-07T16:17:33Z)
• ARNN: Attentive Recurrent Neural Network for Multi-channel EEG Signals to Identify Epileptic Seizures [2.3907933297014927]
  An Attention Recurrent Neural Network (ARNN) is proposed that can process a large amount of data efficiently and accurately. ARNN cell recurrently applies attention layers along a sequence and has linear complexity with the sequence length. This framework is inspired in part by the attention layer and long short-term memory (LSTM) cells, but it scales this typical cell up by several orders to parallelize for multi-channel EEG signals.
  arXiv  Detail & Related papers  (2024-03-05T19:15:17Z)
• DGSD: Dynamical Graph Self-Distillation for EEG-Based Auditory Spatial Attention Detection [49.196182908826565]
  Auditory Attention Detection (AAD) aims to detect target speaker from brain signals in a multi-speaker environment. Current approaches primarily rely on traditional convolutional neural network designed for processing Euclidean data like images. This paper proposes a dynamical graph self-distillation (DGSD) approach for AAD, which does not require speech stimuli as input.
  arXiv  Detail & Related papers  (2023-09-07T13:43:46Z)
• Decoding ECoG signal into 3D hand translation using deep learning [3.20238141000059]
  Motor brain-computer interfaces (BCIs) are promising technology that may enable motor-impaired people to interact with their environment. Most ECoG signal decoders used to predict continuous hand movements are linear models. Deep learning models, which are state-of-the-art in many problems, could be a solution to better capture this relationship.
  arXiv  Detail & Related papers  (2021-10-05T15:41:04Z)
• Transformer-based Spatial-Temporal Feature Learning for EEG Decoding [4.8276709243429]
  We propose a novel EEG decoding method that mainly relies on the attention mechanism. We have reached the level of the state-of-the-art in multi-classification of EEG, with fewer parameters. It has good potential to promote the practicality of brain-computer interface (BCI)
  arXiv  Detail & Related papers  (2021-06-11T00:48:18Z)
• Robust learning from corrupted EEG with dynamic spatial filtering [68.82260713085522]
  Building machine learning models using EEG recorded outside of the laboratory requires robust methods to noisy data and randomly missing channels. We propose dynamic spatial filtering (DSF), a multi-head attention module that can be plugged in before the first layer of a neural network. We tested DSF on public EEG data encompassing 4,000 recordings with simulated channel corruption and on a private dataset of 100 at-home recordings of mobile EEG with natural corruption.
  arXiv  Detail & Related papers  (2021-05-27T02:33:16Z)

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.