NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space Reasoning

Collection of recent arXiv submissions (Mar 19, 2026) focused on electrophysiology ML (EEG→text foundation model, EEG forecasting, EEG-graph explainability, EEG→LLM interfaces), ambulatory ECG cloud platform, and RF/localization/physical-layer advances (WiSLAT Wi‑Fi simultaneous localization & tracking, HRIS self-localization and target positioning, FAS physical-layer secrecy, CRB-based resource allocation). Notable assets and claims: NeuroNarrator plus NeuroCorpus-160K (160k paired EEG segments and clinical text); DECODE and diffusion conditioning on language for event-specific EEG generation