New Foundation Model 'Hypnos' Uses Next-Token Prediction to Learn Sleep Physiology Representations
Researchers have developed Hypnos, a multi-modal foundation model trained on over 20,000 polysomnography recordings using next-token prediction to learn representations of sleep physiology from eight sensing modalities including EEG and ECG. The approach outperforms existing foundation models and matches supervised baselines on sleep stage classification while using 100 times less labeled data. The findings suggest next-token prediction is an effective self-supervised learning objective for physiological signal analysis with potential applications across sleep medicine, cardiology, and neurology.
Researchers have introduced Hypnos, a foundation model designed to compress multi-modal physiological signals into compact health representations. The model was trained on over 20,000 overnight polysomnography recordings using eight different sensing modalities including EEG, ECG, and respiratory signals. Rather than using masked reconstruction or contrastive learning approaches, the team employed next-token prediction—an auto-regressive approach where a large RQ-Transformer jointly predicts the next token across all modalities in parallel. Each modality was first tokenized into discrete streams using residual vector quantization. The model demonstrates significant performance improvements across benchmarks, matching strong supervised baselines on sleep stage classification while requiring 100 times less labeled data. Notably, Hypnos also generalizes beyond sleep to daytime physiology, surpassing a dedicated ECG foundation model at detecting atrial fibrillation, suggesting broad applicability across healthcare domains.
What's missing
The paper does not discuss potential limitations of the approach, such as generalization to different patient populations, equipment variations, or clinical validation requirements before deployment in real-world healthcare settings. Additionally, computational requirements and inference latency are not addressed.
What different sources said
- arXiv cs.AICenter
Next-Token Prediction Learns Generalisable Representations of Sleep Physiology
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.