BrainStack: Neuro-MoE with Functionally Guided Expert Routing for EEG-Based Language Decoding

2026-01-29

AI Summary

This work addresses the challenges posed by the brain’s distributed and nonlinear organization in EEG-based language decoding by proposing the BrainStack framework. Integrating neuroscientific priors with adaptive modeling, BrainStack constructs region-specific expert networks grounded in functional parcellation alongside a global Transformer expert, coordinated through a learnable routing gate that enables context-adaptive collaboration among experts. A top-down cross-regional knowledge distillation mechanism is further introduced to unify functionally guided representation learning with dynamic routing, thereby enhancing both model interpretability and generalization. Experiments on a newly curated large-scale SS-EEG dataset demonstrate that BrainStack significantly outperforms existing methods in cross-subject language decoding, achieving superior accuracy and generalization performance.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Decoding linguistic information from electroencephalography (EEG) remains challenging due to the brain's distributed and nonlinear organization. We present BrainStack, a functionally guided neuro-mixture-of-experts (Neuro-MoE) framework that models the brain's modular functional architecture through anatomically partitioned expert networks. Each functional region is represented by a specialized expert that learns localized neural dynamics, while a transformer-based global expert captures cross-regional dependencies. A learnable routing gate adaptively aggregates these heterogeneous experts, enabling context-dependent expert coordination and selective fusion. To promote coherent representation across the hierarchy, we introduce cross-regional distillation, where the global expert provides top-down regularization to the regional experts. We further release SilentSpeech-EEG (SS-EEG), a large-scale benchmark comprising over 120 hours of EEG recordings from 12 subjects performing 24 silent words, the largest dataset of its kind. Experiments demonstrate that BrainStack consistently outperforms state-of-the-art models, achieving superior accuracy and generalization across subjects. Our results establish BrainStack as a functionally modular, neuro-inspired MoE paradigm that unifies neuroscientific priors with adaptive expert routing, paving the way for scalable and interpretable brain-language decoding.

Problem

• EEG-based language decoding
• brain modularity
• neural decoding
• silent speech
• functional brain organization

Innovation

• Neuro-MoE
• functionally guided routing
• modular brain architecture
• cross-regional distillation
• EEG-based language decoding