Cross-encoder ridge bridge audits detect spectral attribute leakage in frozen EEG embeddings with 95% CI lower bound >=0.081 across model pairs, outperforming single audits like membership inference.
arXiv preprint arXiv:2509.20454 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
JET is a conditional flow matching framework that generates EEG as continuous raw sequences with added constraints for spectral and temporal properties, achieving over 40% lower TS-FID than prior discrete denoising methods on three benchmarks.
citing papers explorer
-
Pretrained, Frozen, Still Leaking: Auditing Cross-Encoder Attribute Transfer in EEG Foundation Models
Cross-encoder ridge bridge audits detect spectral attribute leakage in frozen EEG embeddings with 95% CI lower bound >=0.081 across model pairs, outperforming single audits like membership inference.
-
Let EEG Models Learn EEG
JET is a conditional flow matching framework that generates EEG as continuous raw sequences with added constraints for spectral and temporal properties, achieving over 40% lower TS-FID than prior discrete denoising methods on three benchmarks.