NTU's PhysioWave-T: wavelet biosignal foundation models cut latency 4-7x

NanyangTechnologicalUniversity · hf · 2026-11-10

NTU researchers present PhysioWave-T, a token-efficient wavelet foundation model family for EEG, ECG, and sEMG.

• Dynamic scale folding recombines J+1 wavelet subbands before tokenization, shrinking sequences from (J+1)CS to CS tokens
• Route-specific front ends and learned channel-name embeddings let one Transformer per modality handle diverse montages/sampling rates; unseen setups reuse the backbone via lightweight adapters
• Pre-trained a 30M-param EEG model on 15M+ windows (19–128 channels), plus ECG (30M) and sEMG (12M) encoders
• Folding cuts Transformer latency 4.0–7.2x and peak memory 6.8–12.7x with no accuracy loss; the EEG encoder leads 8 of 9 metrics vs BENDR, BIOT, LaBraM, EEGPT
• RALF fusion raises SeizeIT2 event sensitivity at 11 false alarms/h from 61.4% to 74.4% and window AUPRC from 0.321 to 0.399; SOTA sEMG accuracy and top macro-F1 on PTB-XL Super and Chapman-Shaoxing

Code and models are open-sourced on GitHub.

Original post →

More from Research

• Preprint: Learning Tool-Output Projections from Agent Traces Cuts Billed Tokens up to 56% — solyarisoftware · 2026-11-10
• Shanghai AI Lab's Sage-Taiyi tops AI co-scientist benchmark, lifts reaction yields to 84% — Shanghai-AI-Laboratory · 2026-11-10
• SIRIS benchmark: LLMs can't reliably tell when simulator errors invalidate design recommendations — MicrosoftResearch · 2026-11-10
• How many exposures does a world model need? WARC learns unseen game mechanics in ~80 exposures — third-intelligence · 2026-11-10
• Tencent Hunyuan's FewStepRL adapts RL fine-tuning to few-step diffusion models — Tencent-Hunyuan · 2026-11-10
• BioEmu2 simulations show ligand-triggered conformational shifts in maltose binding protein — FrankNoeBerlin · 2026-11-10

Research channel →