STEAM: A Spatio-TEmporal Alignment Mixture-of-Experts Model with Hierarchical Pre-training for EEG Decoding

• URL: http://arxiv.org/abs/2608.02070v2
• Date: Wed, 05 Aug 2026 06:01:26 GMT
• Title: STEAM: A Spatio-TEmporal Alignment Mixture-of-Experts Model with Hierarchical Pre-training for EEG Decoding
• Abstract summary: We present STEAM, a hierarchical transfer framework that reconciles general-purpose representation learning with paradigm-specific specialization in EEG foundation models.<n> STEAM attains the best average rank among the compared methods at a competitive inference cost measured in FLOPs.
• Score: 12.49365160717963
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Brain-computer interfaces (BCIs) have been widely used in motor rehabilitation, disease diagnosis, and other neural engineering scenarios. However, conventional neural signal decoding algorithms often suffer from limited generalizability and high adaptation costs, motivating recent interest in BCI foundation models. Existing approaches still struggle to jointly achieve general transferability, accurate decoding, and efficient downstream adaptation. We present STEAM, a hierarchical transfer framework that reconciles general-purpose representation learning with paradigm-specific specialization in EEG foundation models. The framework is instantiated as a dual-branch spatio-temporal encoder in which a shared soft mixture-of-experts (SSMoE) module aligns the spatial and temporal branches, allowing complementary representations to exchange information through a compact set of soft slots. Across seven downstream datasets and fourteen evaluation settings, STEAM attains the best average rank among the compared methods at a competitive inference cost measured in FLOPs. Building upon the Stage-I general initialization, the hierarchical pre-training strategy further specializes the model to a target paradigm without retraining from scratch, yielding consistent gains in paradigm-specific decoding accuracy.

Related papers

• CoCoT-EEG: Contrastive-Pretrained Multiscale Convolutional Transformer for EEG Decoding [5.179334032083694]
  We develop a novel contrastive-pretrained EEG model with multiscale temporal convolution input layers and Transformer encoder blocks (CoCoT)<n>CoCoT matches or beats state-of-the-art reconstruction-pretrained EEG models on extensive benchmark decoding tasks with heterogeneous electrode configurations.<n>We demonstrate the viability of contrastive learning for building EEG FMs while suggesting key architectural design considerations.
  arXiv  Detail & Related papers  (2026-07-10T15:48:15Z)
• JSCGC: Joint Source-Channel-Generation Coding for Wireless Generative Communications [58.67925548779465]
  Joint Source-Channel-Generation Coding (JSCGC) is a generative communication paradigm that replaces the conventional decoder with a generative model at the receiver.<n>JSCGC consistently improves feature-based, semantic-level, and inferenceal quality across diverse channel conditions.
  arXiv  Detail & Related papers  (2026-06-11T03:42:03Z)
• MultiDiffNet: A Multi-Objective Diffusion Framework for Generalizable Brain Decoding [1.6528632644902828]
  We introduce textitMultiDiffNet, a diffusion-based framework that bypasses generative augmentation entirely by learning a compact latent space optimized for multiple objectives.<n>We decode directly from this space and achieve state-of-the-art generalization across various neural decoding tasks using subject and session disjoint evaluation.
  arXiv  Detail & Related papers  (2025-11-23T05:22:27Z)
• Explicit modelling of subject dependency in BCI decoding [12.17288254938554]
  Brain-Computer Interfaces (BCIs) suffer from high inter-subject variability and limited labeled data.<n>We present an end-to-end approach that explicitly models the subject dependency using lightweight convolutional neural networks (CNNs) conditioned on the subject's identity.
  arXiv  Detail & Related papers  (2025-09-27T10:51:42Z)
• Towards Efficient General Feature Prediction in Masked Skeleton Modeling [59.46799426434277]
  We propose a novel General Feature Prediction framework (GFP) for efficient mask skeleton modeling.<n>Our key innovation is replacing conventional low-level reconstruction with high-level feature prediction that spans from local motion patterns to global semantic representations.
  arXiv  Detail & Related papers  (2025-09-03T18:05:02Z)
• Zero-Shot EEG-to-Gait Decoding via Phase-Aware Representation Learning [9.49131859415923]
  We propose NeuroDyGait, a domain-generalizable EEG-to-motion decoding framework.<n>It uses structured contrastive representation learning and relational domain modeling to achieve semantic alignment between EEG and motion embeddings.<n>It achieves zero-shot motion prediction for unseen individuals without requiring adaptation and superior performance in cross-subject gait decoding on benchmark datasets.
  arXiv  Detail & Related papers  (2025-06-24T06:03:49Z)
• NDCG-Consistent Softmax Approximation with Accelerated Convergence [67.10365329542365]
  We propose novel loss formulations that align directly with ranking metrics.<n>We integrate the proposed RG losses with the highly efficient Alternating Least Squares (ALS) optimization method.<n> Empirical evaluations on real-world datasets demonstrate that our approach achieves comparable or superior ranking performance.
  arXiv  Detail & Related papers  (2025-06-11T06:59:17Z)
• Deep Unrolled Meta-Learning for Multi-Coil and Multi-Modality MRI with Adaptive Optimization [0.0]
  We propose a unified deep meta-learning framework for accelerated magnetic resonance imaging (MRI)<n>We jointly address multi-coil reconstruction and cross-modality synthesis.<n>Our results show significant improvements in PSNR and over conventional supervised learning.
  arXiv  Detail & Related papers  (2025-05-08T04:47:12Z)
• Model Inversion Attacks Through Target-Specific Conditional Diffusion Models [54.69008212790426]
  Model inversion attacks (MIAs) aim to reconstruct private images from a target classifier's training set, thereby raising privacy concerns in AI applications. Previous GAN-based MIAs tend to suffer from inferior generative fidelity due to GAN's inherent flaws and biased optimization within latent space. We propose Diffusion-based Model Inversion (Diff-MI) attacks to alleviate these issues.
  arXiv  Detail & Related papers  (2024-07-16T06:38:49Z)
• End-to-End Meta-Bayesian Optimisation with Transformer Neural Processes [52.818579746354665]
  This paper proposes the first end-to-end differentiable meta-BO framework that generalises neural processes to learn acquisition functions via transformer architectures. We enable this end-to-end framework with reinforcement learning (RL) to tackle the lack of labelled acquisition data.
  arXiv  Detail & Related papers  (2023-05-25T10:58:46Z)
• Model-based Deep Learning Receiver Design for Rate-Splitting Multiple Access [65.21117658030235]
  This work proposes a novel design for a practical RSMA receiver based on model-based deep learning (MBDL) methods. The MBDL receiver is evaluated in terms of uncoded Symbol Error Rate (SER), throughput performance through Link-Level Simulations (LLS) and average training overhead. Results reveal that the MBDL outperforms by a significant margin the SIC receiver with imperfect CSIR.
  arXiv  Detail & Related papers  (2022-05-02T12:23:55Z)

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.