# SpecMoE: Spectral Mixture-of-Experts Foundation Model for Cross-Species EEG Decoding

arXiv: https://arxiv.org/abs/2603.16739 · HTML: https://mlsanity.com/paper/2603.16739/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Medical and biomedical MoE, one of 12 directions in Mixture-of-experts.
- Builds on 22 of its 36 references that are in the map.
- First posted on arXiv 17 Mar 2026.

## Placed in

- Mixture-of-experts / [Medical and biomedical MoE](https://mlsanity.com/moe/medical-and-biomedical-moe/)

## Builds on (most cited first, 22 of 22)

- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfaces](https://mlsanity.com/paper/1611.08024/) (2016)
- [Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCI](https://mlsanity.com/paper/2405.18765/) (2024)
- [CBraMod: A Criss-Cross Brain Foundation Model for EEG Decoding](https://mlsanity.com/paper/2412.07236/) (2024)
- [REVE: A Foundation Model for EEG -- Adapting to Any Setup with Large-Scale Pretraining on 25,000 Subjects](https://mlsanity.com/paper/2510.21585/) (2025)
- [CSBrain: A Cross-scale Spatiotemporal Brain Foundation Model for EEG Decoding](https://mlsanity.com/paper/2506.23075/) (2025)
- [A Multimodal Approach to Estimating Vigilance Using EEG and Forehead EOG](https://mlsanity.com/paper/1611.08492/) (2016)
- [EEG Foundation Models: A Critical Review of Current Progress and Future Directions](https://mlsanity.com/paper/2507.11783/) (2025)
- [BrainWave: A Brain Signal Foundation Model for Clinical Applications](https://mlsanity.com/paper/2402.10251/) (2024)
- [EEG-FM-Bench: A Comprehensive Benchmark for the Systematic Evaluation and Diagnostic Analyses of EEG Foundation Models](https://mlsanity.com/paper/2508.17742/) (2025)
- [Tokenizing Single-Channel EEG with Time-Frequency Motif Learning](https://mlsanity.com/paper/2502.16060/) (2025)
- [Neural Codecs as Biosignal Tokenizers](https://mlsanity.com/paper/2510.09095/) (2025)
- [EEG-FM-Compass: Progress, Benchmarking, and Future Directions for EEG Foundation Models](https://mlsanity.com/paper/2601.17883/) (2026)
- [Uni-NTFM: A Unified Foundation Model for EEG Signal Representation Learning](https://mlsanity.com/paper/2509.24222/) (2025)
- [LEAD: An EEG Foundation Model for Alzheimer's Disease Detection](https://mlsanity.com/paper/2502.01678/) (2025)
- [DeeperBrain: A Neuro-Grounded EEG Foundation Model Towards Universal BCI](https://mlsanity.com/paper/2601.06134/) (2026)
- [ECHO: Toward Contextual Seq2Seq Paradigms in Large EEG Models](https://mlsanity.com/paper/2509.22556/) (2025)
- [Brain4FMs: A Benchmark of Foundation Models for Electrical Brain Signal](https://mlsanity.com/paper/2602.11558/) (2026)
- [ALFEE: Adaptive Large Foundation Model for EEG Representation](https://mlsanity.com/paper/2505.06291/) (2025)
- [Benchmarking ERP Analysis: Manual Features, Deep Learning, and Foundation Models](https://mlsanity.com/paper/2601.00573/) (2026)
- [CoSupFormer : A Contrastive Supervised learning approach for EEG signal Classification](https://mlsanity.com/paper/2509.20489/) (2025)
- [Aligning Shared and Routed Experts for Cross-Subject EEG Generalization](https://mlsanity.com/paper/2602.01728/) (2026)

## Built on by (most cited first, 1 of 1)

- [MANAS-2: Constrained Reconstruction for EEG Foundation Models](https://mlsanity.com/paper/2609.13717/) (2026)

## Abstract (arXiv)

Decoding the orchestration of neural activity in electroencephalography (EEG) signals is a central challenge in bridging neuroscience with artificial intelligence. Foundation models have made strides in generalized EEG decoding, yet many existing frameworks primarily relying on separate temporal and spectral masking of raw signals during self-supervised pretraining. Such strategies often tend to bias learning toward high-frequency oscillations, as low-frequency rhythmic patterns can be easily inferred from the unmasked signal. We introduce a foundation model that utilizes a novel Gaussian-smoothed masking scheme applied to short-time Fourier transform (STFT) maps. By jointly applying time, frequency, and time-frequency Gaussian masks, we make the reconstruction task much more challenging, forcing the model to learn intricate neural patterns across both high- and low-frequency domains. To effectively recover signals under this aggressive masking strategy, we design SpecHi-Net, a U-shaped hierarchical architecture with multiple encoding and decoding stages. To accelerate large-scale pretraining, we partition the data into three subsets, each used to train an independent expert model. We then combine these models through SpecMoE, a mixture of experts framework guided by a learned spectral gating mechanism. SpecMoE achieves state-of-the-art performance across a diverse set of EEG decoding tasks, including sleep staging, emotion recognition, motor imagery classification, abnormal signal detection, and drug effect prediction. Importantly, the model demonstrates strong cross-species and cross-subject generalization, maintaining high accuracy on both human and murine EEG datasets.