Research map / Label-efficient and robust learning / Recognition and long-tail
EEG-X: Toward Device-Agnostic and Noise-Robust Foundation Models for EEG
Navid Mohammadi Foumani, Soheila Ghane, Nam Nguyen, Mahsa Salehi, Geoffrey Webb, Geoffrey Mackellar
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes EEG-X, a device-agnostic and noise-robust foundation model for EEG aimed at handling electrode/device variability and low signal-to-noise ratio.
Problem
EEG foundation models are immature due to cross-dataset variability from differing recording devices and electrode configurations, and the low SNR of EEG where neural activity is buried under artifacts and non-brain sources.
Approach
A foundation model for EEG designed to be device-agnostic (robust to electrode configurations) and noise-robust; the abstract is truncated so the specific mechanism is not stated.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add location-based channel embedding instead of none (ablation) | EEG-X · cross-domain average (Crowdsourced, BCICIV_2a, SSVEP) | accuracy: 40.02 -> 57.17 on SSVEP; average drop 8.81 without it | Improved |
| use location-based instead of learnable channel embedding (ablation) | EEG-X · cross-domain SSVEP | balanced accuracy: 41.2 -> 57.17 | Improved |
| use location-based instead of learnable channel embedding (ablation) | EEG-X · in-domain average | accuracy: drop of 0.43 with learnable CE vs EEG-X average 77.66 | Improved |
| use location-based instead of learnable channel embedding (ablation) | EEG-X · TUEV in-domain | accuracy: 90.46 (no CE) / 90.14 (learnable CE) -> 91.15 | Mixed |
| add DiCT layer before MSE reconstruction loss (ablation) | EEG-X · in-domain STEW | accuracy: 76.91 -> 80.32 | Improved |
| add DiCT layer before MSE reconstruction loss (ablation) | EEG-X · cross-domain SSVEP | balanced accuracy: 51.55 -> 57.17 | Improved |
| reconstruct ICA-cleaned signal instead of raw signal (ablation) | EEG-X · in-domain STEW | accuracy: 77.90 -> 80.32 | Improved |
| reconstruct ICA-cleaned signal instead of raw signal (ablation) | EEG-X · cross-domain BCICIV_2a | balanced accuracy: 35.31 -> 37.58 | Improved |
| add raw-space reconstruction alongside latent-space reconstruction (ablation) | EEG-X · in-domain STEW | accuracy: 75.92 -> 80.32 | Improved |
| use ICA as preprocessing instead of raw input | EEG-X · STEW | accuracy: 75.12 -> 70.68 | Hurt |
| use ICA as preprocessing instead of raw input | EEG-X · TUEV | accuracy: 88.24 -> 83.36 | Hurt |
| use ICA as preprocessing instead of raw input | EEG-X · DREAMER and TUAB | accuracy: 55.69 -> 56.24 (DREAMER); 78.12 -> 78.25 (TUAB) | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- EEGPT — EEGPT: Pretrained Transformer for Universal and Reliable Representation of EEG Signals (NeurIPS 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- What do EEG Foundation Models Capture from Human Brain Signals? (NeurIPS 2026 · similarity 0.90)
- NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces (NeurIPS 2026 · similarity 0.89)
- CBraMod: A Criss-Cross Brain Foundation Model for EEG Decoding (ICLR 2025 · similarity 0.89)
- BrainOmni: A Brain Foundation Model for Unified EEG and MEG Signals (NeurIPS 2025 · similarity 0.89)
- Standing on the Shoulders of Giants: Rethinking EEG Foundation Model Pretraining via Multi-Teacher Distillation (NeurIPS 2026 · similarity 0.88)
- NeurIPT: Foundation Model for Neural Interfaces (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
