Controlled Evaluation of EEG Foundation Models: Random Initialisation, Band-Power and Validation-Selected Probes

The paper evaluates five publicly released EEG foundation models and an in‑house JEPA‑style model using a protocol that includes three key controls: an untrained encoder, spectral band‑power features, and a probe layer selected without test‑set look‑ahead. The study finds that selecting the probe layer by test score can modestly improve peak performance and alter the peak‑versus‑output relationship, that random initialization can match or surpass pretraining on certain tasks, and that full fine‑tuning rarely yields gains beyond seed variability. These findings lead the authors to propose four new reporting practices and to release all experimental results for re‑analysis.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

• fine-tuning
• benchmarks