Reproduce “ERP-FM: A Foundation Model for Universal ERP Analysis” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yihe Wang, Bohan Chen, Taida Li, Yujun Yan, Rui Yin, Xiang Zhang
ERP-FM pretrains on individual event-linked EEG trials for downstream event and disease classification. It divides electrode signals into time patches and learns to reconstruct masked patches, then adapts the encoder using single or averaged trials. Across 36 dataset–metric evaluations, it achieved rank 2.88 and ranked first in 20. On RLPD Parkinson’s-versus-healthy-control classification, it reached 94.97% subject-level F1, a dataset-specific result.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.