EEG-FM-Audit is an evaluation pipeline showing that properly tuned supervised baselines can match or outperform EEG foundation models with far fewer parameters on public datasets.
2601.17883 , archivePrefix=
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 7roles
background 1polarities
background 1representative citing papers
Reconstruction-based EEG foundation models preferentially encode aperiodic and low-frequency components over oscillatory structure, with embeddings capturing subject identity more than task-relevant information.
NeuroAtlas benchmarks foundation models on 42 EEG datasets and reports that EEG-specific models do not consistently outperform generic time-series models, standard metrics miss clinical utility, and rankings vary by domain.
NeuralBench is a new benchmarking framework for neuroAI models on EEG data that finds foundation models only marginally outperform task-specific ones while many tasks like cognitive decoding stay highly challenging.
FUSED integrates EEG foundation models into source-free domain adaptation via dual-branch co-adaptation, consensus filtering, and two-stage pseudo-label refinement to achieve state-of-the-art cross-subject EEG decoding.
MST combines Morlet wavelet tokenization, long-context baseline removal, and frequency-specific spatial projections with a Transformer backbone to outperform pretrained EEG models on SEED datasets for cross-subject emotion decoding.
LRP on EEG transformers reveals Clever Hans artifacts in motor imagery tasks and a recurring central electrode cluster as a candidate sensorimotor signature of arousal.
citing papers explorer
-
EEG-FM-Audit: A Systematic Evaluation and Analysis Pipeline for EEG Foundation Models
EEG-FM-Audit is an evaluation pipeline showing that properly tuned supervised baselines can match or outperform EEG foundation models with far fewer parameters on public datasets.
-
Aperiodic and Low-Frequency Spectral Bias in Reconstruction based EEG Foundation Models
Reconstruction-based EEG foundation models preferentially encode aperiodic and low-frequency components over oscillatory structure, with embeddings capturing subject identity more than task-relevant information.
-
NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces
NeuroAtlas benchmarks foundation models on 42 EEG datasets and reports that EEG-specific models do not consistently outperform generic time-series models, standard metrics miss clinical utility, and rankings vary by domain.
-
NeuralBench: A Unifying Framework to Benchmark NeuroAI Models
NeuralBench is a new benchmarking framework for neuroAI models on EEG data that finds foundation models only marginally outperform task-specific ones while many tasks like cognitive decoding stay highly challenging.
-
Foundation Model Guided Dual-Branch Co-Adaptation for Source-Free EEG Decoding
FUSED integrates EEG foundation models into source-free domain adaptation via dual-branch co-adaptation, consensus filtering, and two-stage pseudo-label refinement to achieve state-of-the-art cross-subject EEG decoding.
-
Dive into Waves: Morlet Spectral Transformer for Cross-Subject Emotion Decoding from EEG
MST combines Morlet wavelet tokenization, long-context baseline removal, and frequency-specific spatial projections with a Transformer backbone to outperform pretrained EEG models on SEED datasets for cross-subject emotion decoding.
-
From Clever Hans to Scientific Discovery: Interpreting EEG Foundational Transformers with LRP
LRP on EEG transformers reveals Clever Hans artifacts in motor imagery tasks and a recurring central electrode cluster as a candidate sensorimotor signature of arousal.