NeuroNarrator Translates EEG Signals Into Narratives

On Feb 24, 2026, researchers led by Guoan Wang present NeuroNarrator, a generalist EEG-to-text foundation model trained on NeuroCorpus-160K pairing over 160,000 EEG segments with structured clinical descriptions. The architecture aligns temporal waveforms with spatial topographic maps via a contrastive objective and conditions a large language model using a state-space-inspired temporal-spectral context. Evaluations report strong performance across diverse benchmarks and zero-shot transfer tasks.
Scoring Rationale
Novel EEG-to-text foundation and large dataset drive score, limited by preprint status and niche clinical scope.
Practice with real Health & Insurance data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Health & Insurance problems

