Research map / Architectures and efficiency / Pruning and efficient tuning
Are EEG Foundation Models Worth It? Comparative Evaluation with Traditional Decoders in Diverse BCI Tasks
Liuyin Yang, Qiang Sun, Ang Li, Marc M. Van Hulle
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune instead of linear probe the foundation model (ablation) | EEG foundation models · Pooled classification tasks | accuracy | Improved |
| use population-trained models instead of per-subject models (ablation) | Foundation models · LOO Zero-Shot vs Per-Subject (Transfer) | cross-subject accuracy | Improved |
| subject-specific fine-tuning instead of linear probing (ablation) | Foundation models · LOO Drop (population subjects) | accuracy drop | Hurt |
| fine-tuned foundation models vs classic NN and non-NN decoders | ST-EEGFormer-l and fine-tuned foundation models · Population decoding | accuracy | Improved |
| foundation models vs classic decoders in per-subject and LOO zero-shot settings | Foundation models · Per-Subject (Self), LOO Zero-Shot | accuracy | No clear effect |
| linear-probed foundation models vs classic NN decoders (CTNet) | Linear-probed foundation models · Most evaluation protocols | aggregated rank | Hurt |
| scale up model size across all NN decoders (ablation) | NN-based EEG decoders · Downstream EEG classification | normalized accuracy | No clear effect |
| scale ST-EEGFormer from small/base to large (ablation) | ST-EEGFormer · Classification tasks | accuracy | Improved |
| fine-tune foundation models for regression instead of linear probe (ablation) | Foundation models · DTU regression | Pearson R: LaBraM 0.011 (f) vs 0.024 (l) | Hurt |
| use multi-layer head or full-token fusion instead of single linear head with average-token fusion (ablation) | Foundation models (linear probing) · Classification | accuracy | Improved |
Closest papers by meaning
- What do EEG Foundation Models Capture from Human Brain Signals? (NeurIPS 2026 · similarity 0.91)
- CBraMod: A Criss-Cross Brain Foundation Model for EEG Decoding (ICLR 2025 · similarity 0.91)
- Standing on the Shoulders of Giants: Rethinking EEG Foundation Model Pretraining via Multi-Teacher Distillation (NeurIPS 2026 · similarity 0.91)
- NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces (NeurIPS 2026 · similarity 0.91)
- NeurIPT: Foundation Model for Neural Interfaces (NeurIPS 2025 · similarity 0.90)
- CSBrain: A Cross-scale Spatiotemporal Brain Foundation Model for EEG Decoding (NeurIPS 2025 · similarity 0.90)
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.
