# Channel Adaptation for EEG Foundation Models: A Systematic Benchmark Across Architectures, Tasks, and Training Regimes

arXiv: https://arxiv.org/abs/2604.23091 · HTML: https://mlsanity.com/paper/2604.23091/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 25 references that are in the map.
- First posted on arXiv 25 Apr 2026.

## Builds on (most cited first, 7 of 7)

- [BENDR: using transformers and a contrastive self-supervised learning task to learn from massive amounts of EEG data](https://mlsanity.com/paper/2101.12037/) (2021)
- [CBraMod: A Criss-Cross Brain Foundation Model for EEG Decoding](https://mlsanity.com/paper/2412.07236/) (2024)
- [Uncovering the structure of clinical EEG signals with self-supervised learning](https://mlsanity.com/paper/2007.16104/) (2020)
- [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)
- [Neuro-GPT: Towards A Foundation Model for EEG](https://mlsanity.com/paper/2311.03764/) (2023)
- [EEG Foundation Challenge: From Cross-Task to Cross-Subject EEG Decoding](https://mlsanity.com/paper/2506.19141/) (2025)
- [Physics-informed and Unsupervised Riemannian Domain Adaptation for Machine Learning on Heterogeneous EEG Datasets](https://mlsanity.com/paper/2403.15415/) (2024)

## Built on by (most cited first, 5 of 5)

- [ReMAP: Self-supervised learning to unveil brain representations and vulnerability](https://mlsanity.com/paper/2608.22042/) (2026)
- [Separating personal from population gains when calibrating EEG foundation models for new users](https://mlsanity.com/paper/2609.34801/) (2026)
- [Representational and Functional Robustness to Electrode Montages in EEG Foundation Models](https://mlsanity.com/paper/2609.36288/) (2026)
- [EMG-FM-Bench: A Comprehensive Benchmark for Foundation Model Transfer and Adaptation on Electromyography](https://mlsanity.com/paper/2610.06450/) (2026)
- [HyBRA-EEG: Hyperbolic Global-Local Brain-Region Aggregation for Self-Supervised EEG Representation Learning](https://mlsanity.com/paper/2610.16589/) (2026)

## Abstract (arXiv)

Scaling EEG foundation models requires pooling data across heterogeneous electrode montages, a prerequisite both for larger pretraining corpora and for downstream deployment. We present the first systematic comparison of four channel adaptation methods (Conv1d projection, spherical spline interpolation (SSI), source-space decomposition, and Riemannian re-centering) across five pretrained EEG foundation models (5M--157M parameters), five downstream tasks, and two training regimes with 10--15 random seeds each. We find that rigid-montage models (BENDR, Neuro-GPT) require external adaptation, while flexible models (EEGPT, CBraMod) match or exceed it natively when fine-tuned but benefit from external methods under frozen-encoder deployment. A probe-SFT asymmetry exists: external adaptation can cause severe negative transfer during fine-tuning of flexible models. The optimal method is architecture-dependent (Conv1d for BENDR, SSI/Riemannian for Neuro-GPT, source-space decomposition for depression detection), and 5M-parameter CBraMod outperforms models up to 31× larger on 4/5 datasets, consistent with independent findings that compact EEG-specific architectures can match larger models.