Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

OSF: On Pre-training and Scaling of Sleep Foundation Models

T0 review · 3 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Channel masking during pre-training makes sleep foundation models robust to missing sensors.

desk verdict The benchmark and channel-masking result are solid, but the multi-source scaling claim is confounded by sample size; worth refereeing with that fixed. read the letter →

arxiv 2603.00190 v2 pith:XHPF2IT4 submitted 2026-02-27 cs.LG cs.AI

Zitao Shuai , Zongzhe Xu , David Yang , Wei Wang , Yuzhe Yang This is my paper
classification cs.LGcs.AI
keywords sleepfoundationmodelspolysomnographyself-supervisedpre-trainingchannelinvariancemissing-channelrobustnessscalinglawsmulti-sourcedatadiseaseprediction
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that the pre-training recipe determines whether a sleep foundation model generalizes to real clinical conditions, where sensor sets differ and channels drop out. Evaluating four families of self-supervised objectives on a 166,500-hour corpus of polysomnography (multi-signal sleep recordings) from nine public datasets, it finds that existing models lose substantial accuracy when brain or respiratory channels are missing at inference, and that pre-training with random 50% channel masking—zeroing out half of the 12 standardized channels—restores robustness and improves transfer. With that recipe, models keep improving as pre-training data, model capacity, and the number of data sources increase, and the authors package the findings into OSF, which reports top performance across sleep staging, sleep-event detection, and patient-level disease prediction on all nine cohorts. The paper also acknowledges that a meaningful gap remains when task-critical channels are entirely absent, so the recipe narrows rather than eliminates the missing-channel problem.

What carries the argument

The central object is OSF, a pre-training recipe built on self-distillation with a two-stage masking augmentation: randomly zero 50% of the input channels, then apply contiguous temporal block masking to the remaining signals. A Transformer encoder is trained on 30-second epochs, resampled to 64 Hz from a standardized 12-channel montage (EEG, EOG, respiratory effort, airflow, snore, ECG, and three EMG leads). The channel-masking step carries the argument: by making two views of the same epoch differ in which physiological signals are present, the encoder cannot rely on sensor identity and must learn features aligned with physiology. For patient-level tasks, a top-k selection module aggregate

What would settle it

Evaluate an OSF checkpoint on genuine recordings that never collected a particular sensor (for example, a home study without EEG) and compare performance to the paper's zero-masking simulation on full PSG recordings; if the real missing-sensor gap is much larger than the simulated gap, the channel-invariance conclusion does not transfer to practice.

Watch

Extended reading notes

Core claim

At the center is a negative result and a fix. The negative result: current sleep foundation models are not reliable when inference-time inputs lack channels they saw during pre-training—removing brain channels hurts sleep staging, removing respiratory channels hurts hypopnea detection. The fix: explicitly force channel-invariant feature learning during pre-training by randomly zeroing 50% of input channels and requiring two augmented views to align. The authors show this works for both contrastive and self-distillation objectives, that it turns data scaling from a saturating curve into a consistently improving one, and that the resulting OSF family yields the strongest representations in lin

Load-bearing premise

The robustness and deployment claims rest on treating zero-masked or zero-padded channels as a faithful stand-in for sensors that were never recorded, even though real dropouts also remove channel identity and calibration information.

Editorial extensions

If this is right

  • Sleep models built this way will tolerate realistic deployment settings—headband-only EEG studies, home studies without brain signals, airflow-only setups—with much smaller accuracy drops than current models.
  • Pooling many public sleep cohorts is worth the distribution shift: multi-source pre-training improves out-of-domain generalization at every data scale tested.
  • Scaling continues to pay off: larger transformers and more pre-training hours turn into better linear-probing, fine-tuning, and few-shot performance, giving a concrete roadmap to stronger sleep models.
  • Rare sleep-event labels can be exploited more efficiently; OSF is consistently better with 1, 5, and 50 labeled examples per class, which matters for tasks with scarce annotations.
  • Patient-level disease prediction should select informative epochs rather than average the whole night; a top-k selector outperforms mean pooling and trainable sequence aggregators in the paper's comparisons.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The missing-channel story is tested by zero-masking recorded signals; a direct comparison on recordings where sensors were physically absent would separate true channel invariance from learned zero-pattern shortcuts, and is the natural next experiment.
  • If channel dropout is the mechanism, the same augmentation should transfer to other physiological time-series models—wearable ECG, respiratory bands, home monitors—where sensor absence is common; the paper's recipe is directly testable there.
  • The top-k aggregation finding suggests disease-relevant information concentrates in specific sleep epochs; a clinical follow-up could check whether the selected epochs correspond to interpretable stages or events like apnea or arousal.
  • The scaling curves are demonstrated up to 85M parameters and roughly 166,500 hours; whether they flatten at substantially larger scale—or whether data-hungry objectives like OSF's will dominate—are questions the paper leaves open.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper introduces SleepBench, a large open benchmark of 166,500 hours of polysomnography from nine public datasets, and uses it to systematically compare four families of self-supervised pre-training objectives (contrastive, reconstruction, autoregressive, distillation) for sleep foundation models. Three findings are claimed: (1) existing sleep FMs degrade sharply when channels are missing at inference; (2) channel-masking augmentation, which encourages channel-invariant representations, substantially improves downstream transfer for invariance-based methods; and (3) scaling pre-training data, model capacity, and multi-source data mixture consistently improves performance. The authors then propose OSF, a DINO-based model with a two-stage time/channel masking pre-training recipe, and report state-of-the-art results across sleep staging, sleep event detection, few-shot adaptation, and patient-level disease prediction on in-domain and out-of-domain cohorts.

Significance. If the findings hold, this is a valuable contribution: it provides the largest fully open sleep benchmark to date, a controlled comparison of SSL design choices in a domain where such analyses are scarce, and a concrete pre-training recipe (channel masking plus self-distillation) that appears to improve transfer and robustness. The strengths include the scale and openness of the corpus, the use of held-out external cohorts (e.g., MROS), and detailed ablations. However, the load-bearing claim that multi-source data diversity helps is confounded with total pre-training sample size, and the missing-channel evaluation relies on zero-padding as a proxy for absent sensors; both issues affect the central findings. The lack of error bars further weakens the quantitative SOTA comparisons. With these points addressed, the paper would be a strong candidate for acceptance.

major comments (3)
  1. [Sec. 5 (Takeaway 1), Appendix D.2, Tables 26-29] The multi-source versus single-source comparison does not hold pre-training sample size fixed. Single-source is SHHS-only (5.9M train epochs at 100%, Table 14), while multi-source uses the full corpus (16.0M train epochs at 100%). At each data percentage (1%, 10%, 100%), multi-source uses ~2.7x more epochs. Since the paper's own Finding 3 (Sec. 3.3, Fig. 5) shows that downstream performance improves with pre-training sample size, the multi-source advantage in Tables 26-29 and Fig. 8 is fully confounded with total epoch count. The text in D.2 says the variants are trained 'under identical settings,' but identical percentage of each cohort is not identical sample size. Therefore, the diversity component of the recipe is unsupported. Please add a matched-sample-size control, e.g., subsample the multi-source corpus to the SHHS epoch count, or train SHHS-only with repeated epochs, to isolate
  2. [Sec. 2, Sec. 3.1, Sec. 6.1 (Tables 5-6)] All missing-channel evaluation uses zero-padded or zero-masked channels as a proxy for absent sensors. The model may exploit zero-pattern artifacts (constant-value channels, padding positions) rather than learn true channel-invariant representations, and real missing channels also remove channel identity/calibration information that the architecture may still receive. The claims 'existing sleep FMs fail to generalize under missing-channel inference' (Finding 1), 'channel-invariant feature learning is essential' (Finding 2), and that OSF 'better addresses practical constraints' (Sec. 6.1) rest on this simulation. Please provide evidence on naturally missing-channel cohorts (e.g., datasets that natively lack certain channels) or analyze sensitivity to the zero-padding implementation, such as random channel dropout at inference or a learned mask embedding. The paper's acknowledgment that 'a
  3. [All main tables (e.g., Tables 1-6, 26-32)] No error bars or multiple seeds are reported. Many decisive comparisons are small: e.g., Table 2 fine-tuning hypopnea AUC (OSF 85.0 vs MAE 85.1), Table 4 diabetes AUPRC (OSF 56.3 vs VQ-VAE 54.6), and Table 6 missing-channel gains often under 1.0 AUC. Since each method was independently tuned to its 'best-performing regime' (App. A.1), it is unclear whether the reported gaps exceed run-to-run noise. Please report mean ± std over at least 3 seeds, or provide paired significance tests for the main OSF-versus-baseline comparisons. Without this, the state-of-the-art claim is not quantitatively grounded.
minor comments (5)
  1. [Sec. 3.2] The masking-ratio specification 'r∈0.3,0.6' should be written as r ~ Uniform[0.3, 0.6]; the interval notation is ambiguous.
  2. [Table 10] The header row 'AROSF(SimCLR)OSF(DINO)' appears to collapse multiple columns; it should be 'AR | OSF(SimCLR) | OSF(DINO)' or similar.
  3. [Table 17] Typographical errors: 'Somantic' should be 'Somatic' and 'Nassal Pressure' should be 'Nasal Pressure'.
  4. [Tables 20-22] The method 'MTAE' appears in few-shot results but is never defined in the main text or appendix; if this is a typo for 'MAE', please correct it, otherwise define it.
  5. [Table 3] The row label 'In-Domain Datasets Out-of-Domain Datasets' is confusing; the table would benefit from a clear separator or a caption indicating which columns correspond to in-domain vs OOD cohorts.

Circularity Check

0 steps flagged · score 0.0 of 10

No constructional circularity; OSF's findings are empirical, controlled pre-training comparisons evaluated on held-out downstream tasks.

full rationale

The paper's central claims are empirical results rather than a derivation chain. OSF is formed by taking DINO and adding channel masking and scaling, and its reported improvements come from linear probing, fine-tuning, few-shot, and held-out OOD evaluations on MROS and other cohorts. Finding 1 is a direct benchmark observation about existing FMs. Finding 2 is an ablation of masking strategies on downstream tasks; it is not defined in terms of the metric it predicts. Finding 3 is an observed scaling trend with pre-training data size, model size, and data mixture. None of these reduce by construction to the recipe's inputs: the pre-training augmentations are not the downstream labels, the benchmark is not the loss being optimized, and no fitted parameter is renamed as a prediction. There is no load-bearing self-citation or imported uniqueness theorem. The single-source vs. multi-source comparison does not hold total epoch count fixed (at matched percentages, multi-source uses roughly 16.0M epochs vs. 5.9M for SHHS-only), and zero-padding/zero-masking may not perfectly mimic real channel absence; however, these are internal-validity and deployment-realism concerns, not circularity. The paper also explicitly acknowledges remaining gaps under missing-channel settings, further indicating the claims are not presented as forced by construction.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central claims depend on domain modeling choices rather than mathematical axioms: epoch-level segmentation, label comparability across cohorts, zero-padding as missing-channel proxy, and a fixed 12-channel montage. These are reasonable but are assumptions imported without independent verification within the paper. OSF is a model family and SleepBench a dataset, not a newly postulated physical entity; no new forces, particles, or conserved quantities are introduced.

free parameters (6)
  • channel-mask drop fraction = 0.5
    Randomly drops 50% of channels in each augmented view; chosen by hand as the channel-invariance augmentation (Sec. 3.2, A.1).
  • temporal masking ratio range = U(0.3, 0.6)
    Sampled per example; augmentation hyperparameter in Sec. 3.2/A.1.
  • cropping ratio range = U(0.25, 0.75)
    Vision-style augmentation compared in Table 8.
  • z-score clipping threshold = [-6, 6]
    Chosen based on empirical distribution of z-scored signals (App. B.3).
  • token/window length = 64
    Input token window length in Table 10; signals resampled to 64 Hz and windowed to 64 tokens.
  • top-k selection count = not reported
    Top-k patient-level aggregator in Table 7; k value is not stated in the paper, so exact reproducibility is incomplete.
assumptions (4)
  • domain assumption 30-second epoch segmentation is the right unit for all downstream tasks, including patient-level disease prediction
    Used throughout Sec. 2/4.1; if disease signals are not captured in 30-s epoch embeddings, the aggregation results do not follow.
  • domain assumption Held-out MROS is a representative OOD cohort and subject-level split prevents leakage
    Sec. 2 and B.2; if MROS distributions differ in ways not captured by the fixed 12-channel montage, OOD conclusions may not generalize.
  • domain assumption Event and disease labels across nine datasets are compatible enough to compare downstream metrics
    Sec. 2/B.2; label definitions and event prevalence differ (Table 15), and no harmonization is described beyond preprocessing.
  • domain assumption Zero-padding missing channels is a faithful simulation of real channel absence
    Sec. 2/3.1; this is the premise enabling all missing-channel experiments; if false, Findings 1-2 may be artifacts.

how reviews work

0 comments
Cite this review

Pith. "Pith review of OSF: On Pre-training and Scaling of Sleep Foundation Models." pith.science (2026). https://pith.science/paper/XHPF2IT4

@misc{pith2026260300190,
  author       = {Pith},
  title        = {Pith review of: OSF: On Pre-training and Scaling of Sleep Foundation Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XHPF2IT4}},
  note         = {Machine review of arXiv:2603.00190}
}
read the original abstract

Polysomnography (PSG) provides the gold standard for sleep assessment but suffers from substantial heterogeneity across recording devices and cohorts. There have been growing efforts to build general-purpose foundation models (FMs) for sleep physiology, but lack an in-depth understanding of the pre-training process and scaling patterns that lead to more generalizable sleep FMs. To fill this gap, we curate a massive corpus of 166,500 hours of sleep recordings from nine public sources and establish SleepBench, a comprehensive, fully open-source benchmark. Leveraging SleepBench, we systematically evaluate four families of self-supervised pre-training objectives and uncover three critical findings: (1) existing FMs fail to generalize to missing channels at inference; (2) channel-invariant feature learning is essential for pre-training; and (3) scaling sample size, model capacity, and multi-source data mixture consistently improves downstream performance.With an enhanced pre-training and scaling recipe, we introduce OSF, a family of sleep FMs that achieves state-of-the-art performance across nine datasets on diverse sleep and disease prediction tasks. Further analysis of OSF also reveals intriguing properties in sample efficiency, hierarchical aggregation, and cross-dataset scaling.

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Inertia-1: An Open Exploration of Wearable Motion Foundation Models

    cs.LG 2026-07 conditional novelty 6.5 of 10

    Controlled large-scale pretraining on 18.2M hours of wearables shows self-supervised motion models beat scratch training, with triaxial fidelity, data diversity, and task-matched windows mattering more than model size alone.

  2. Signal or Noise? Understanding Generative Models for Real-World Sensor Time Series

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Across 14 sensor generation settings, flow-matching models are the strongest overall baseline, while demographic conditioning, time-frequency modeling, and moderate synthetic augmentation improve hard regimes and down...

Reference graph

Works this paper leans on

45 extracted references · 8 linked inside Pith · cited by 2 Pith papers

  1. [1]

    Large-scale training of foundation models for wearable biosignals

    Salar Abbaspourazad, Oussama Elachqar, Andrew Miller, Saba Emrani, Udhyakumar Nallasamy, and Ian Shapiro. Large-scale training of foundation models for wearable biosignals. InThe Twelfth International Conference on Learning Representations

  2. [2]

    Chronos: Learning the language of time series.arXiv preprint arXiv:2403.07815, 2024

    Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, et al. Chronos: Learning the language of time series.arXiv preprint arXiv:2403.07815, 2024

    arXiv 2024

  3. [3]

    The dreem headband compared to polysomnography for electroencephalographic signal acquisition and sleep staging.Sleep, 43(11):zsaa097, 2020

    PierrickJArnal,ValentinThorey,EdenDebellemaniere,MichaelEBallard,AlbertBouHernandez, Antoine Guillot, Hugo Jourde, Mason Harris, Mathias Guillard, Pascal Van Beers, et al. The dreem headband compared to polysomnography for electroencephalographic signal acquisition and sleep staging.Sleep, 43(11):zsaa097, 2020. 11 OSF: On Pre-training and Scaling of Sleep...

    2020

  4. [4]

    Validation of a self- applied unattended monitor for sleep disordered breathing.Journal of Clinical Sleep Medicine, 4(1):26–37, 2008

    Indu Ayappa, Robert G Norman, Vijay Seelall, and David M Rapoport. Validation of a self- applied unattended monitor for sleep disordered breathing.Journal of Clinical Sleep Medicine, 4(1):26–37, 2008

    2008

  5. [5]

    Evaluating reliability in wearable devices for sleep staging.NPJ Digital Medicine, 7(1):74, 2024

    Vera Birrer, Mohamed Elgendi, Olivier Lambercy, and Carlo Menon. Evaluating reliability in wearable devices for sleep staging.NPJ Digital Medicine, 7(1):74, 2024

    2024

  6. [6]

    Terri Blackwell, Kristine Yaffe, Sonia Ancoli-Israel, Susan Redline, Kristine E Ensrud, Marcia L Stefanick, Alison Laffan, Katie L Stone, and Osteoporotic Fractures in Men Study Group. Associations between sleep architecture and sleep-disordered breathing and cognition in older community-dwelling men: the osteoporotic fractures in men sleep study.Journal ...

    2011

  7. [7]

    Emerging properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerging properties in self-supervised vision transformers. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 9650–9660, 2021

    2021

  8. [8]

    wav2sleep: A unified multi-modal approach to sleep stage classification from physiological signals

    Jonathan F Carter and Lionel Tarassenko. wav2sleep: A unified multi-modal approach to sleep stage classification from physiological signals. InMachine Learning for Health (ML4H), pages 186–202. PMLR, 2025

    2025

Show all 45 references
  1. [9]

    A simple framework for contrastive learning of visual representations

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. A simple framework for contrastive learning of visual representations. InInternational conference on machine learning, pages 1597–1607. PmLR, 2020

    2020

  2. [10]

    Racial/ethnic differences in sleep disturbances: the multi-ethnic study of atherosclerosis (mesa).Sleep, 38(6):877–888, 2015

    Xiaoli Chen, Rui Wang, Phyllis Zee, Pamela L Lutsey, Sogol Javaheri, Carmela Alcántara, Chandra L Jackson, Michelle A Williams, and Susan Redline. Racial/ethnic differences in sleep disturbances: the multi-ethnic study of atherosclerosis (mesa).Sleep, 38(6):877–888, 2015

    2015

  3. [11]

    This time is different: An observability perspective on time series foundation models.arXiv preprint arXiv:2505.14766, 2025

    Ben Cohen, Emaad Khwaja, Youssef Doubli, Salahidine Lemaachi, Chris Lettieri, Charles Masson, Hugo Miccinilli, Elise Ramé, Qiqi Ren, Afshin Rostamizadeh, et al. This time is different: An observability perspective on time series foundation models.arXiv preprint arXiv:2505.14766, 2025

    2025

  4. [12]

    An image is worth 16x16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929, 2020

    Alexey Dosovitskiy. An image is worth 16x16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929, 2020

    2010 arXiv

  5. [13]

    Challenges in recruitment to a randomized controlled study of cardiovascular disease reduction in sleep apnea: an analysis of alternative strategies.Sleep, 37(12):2035–2038, 2014

    Kevin Gleason, Donghoon Shin, Michael Rueschman, Tanya Weinstock, Rui Wang, James H Ware, Murray A Mittleman, and Susan Redline. Challenges in recruitment to a randomized controlled study of cardiovascular disease reduction in sleep apnea: an analysis of alternative strategies...

  6. [14]

    Naihua N Gong, Aditya Mahat, Samya Ahmad, Daniel Glaze, Mirjana Maletic-Savatic, Matthew McGinley, Anne Marie Morse, Alcibiades J Rodriguez, Audrey Thurm, Susan Redline, et al. Leveraging clinical sleep data across multiple pediatric cohorts for insights into neurodevelop- men...

    2025

  7. [15]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. Masked autoencoders are scalable vision learners. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 16000–16009, 2022

    2022

  8. [16]

    Long short-term memory.Neural computation, 9(8):1735–1780, 1997

    Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory.Neural computation, 9(8):1735–1780, 1997. 12 OSF: On Pre-training and Scaling of Sleep Foundation Models

    1997

  9. [17]

    Attention-based deep multiple instance learning

    Maximilian Ilse, Jakub Tomczak, and Max Welling. Attention-based deep multiple instance learning. InInternational conference on machine learning, pages 2127–2136. PMLR, 2018

    2018

  10. [18]

    Ziyu Jia, Youfang Lin, Jing Wang, Xiaojun Ning, Yuanlai He, Ronghao Zhou, Yuhan Zhou, and Li-wei H Lehman. Multi-view spatial-temporal graph convolutional networks with domain gen- eralization for sleep stage classification.IEEE Transactions on Neural Systems and Rehabilitatio...

    1977

  11. [19]

    Stanford sleep bench: Evaluating polysomnography pre-training methods for sleep foundation models

    Magnus Ruud Kjaer, Rahul Thapa, Gauri Ganjoo, Hyatt Moore IV, Poul Joergen Jennum, Bran- don M Westover, James Zou, Emmanuel Mignot, Bryan He, and Andreas Brink-Kjaer. Stanford sleep bench: Evaluating polysomnography pre-training methods for sleep foundation models. arXiv prep...

    2025

  12. [20]

    A large collection of real-world pediatric sleep studies.Scientific Data, 9(1):421, 2022

    Harlin Lee, Boyue Li, Shelly DeForte, Mark L Splaingard, Yungui Huang, Yuejie Chi, and Simon L Linwood. A large collection of real-world pediatric sleep studies.Scientific Data, 9(1):421, 2022

    2022

  13. [21]

    Self-supervised contrastive learning for medical time series: A systematic review.Sensors, 23(9):4221, 2023

    Ziyu Liu, Azadeh Alavi, Minyi Li, and Xiang Zhang. Self-supervised contrastive learning for medical time series: A systematic review.Sensors, 23(9):4221, 2023

    2023

  14. [22]

    Decoupled weight decay regularization.arXiv preprint arXiv:1711.05101, 2017

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization.arXiv preprint arXiv:1711.05101, 2017

    2017 arXiv

  15. [23]

    A randomized trial of adenotonsillectomy for childhood sleep apnea.New England Journal of Medicine, 368(25):2366– 2376, 2013

    Carole L Marcus, Reneé H Moore, Carol L Rosen, Bruno Giordani, Susan L Garetz, H Gerry Taylor, Ron B Mitchell, Raouf Amin, Eliot S Katz, Raanan Arens, et al. A randomized trial of adenotonsillectomy for childhood sleep apnea.New England Journal of Medicine, 368(25):2366– 2376, 2013

    2013

  16. [24]

    Umap: Uniform manifold approximation and projection for dimension reduction.arXiv preprint arXiv:1802.03426, 2018

    Leland McInnes, John Healy, and James Melville. Umap: Uniform manifold approximation and projection for dimension reduction.arXiv preprint arXiv:1802.03426, 2018

    2018 arXiv

  17. [25]

    Scaling wearable foundation models.arXiv preprint arXiv:2410.13638, 2024

    Girish Narayanswamy, Xin Liu, Kumar Ayush, Yuzhe Yang, Xuhai Xu, Shun Liao, Jake Garrison, Shyam Tailor, Jake Sunshine, Yun Liu, et al. Scaling wearable foundation models.arXiv preprint arXiv:2410.13638, 2024

    2024 arXiv

  18. [26]

    Dinov2: Learning robust visual features without supervision.arXiv preprint arXiv:2304.07193, 2023

    Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy Vo, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision.arXiv preprint arXiv:2304.07193, 2023

    2023 arXiv

  19. [27]

    Pedsleepmae: Generativemodelformultimodal pediatric sleep signals

    SauravRajPandey, AaqibSaeed, andHarlinLee. Pedsleepmae: Generativemodelformultimodal pediatric sleep signals. In2024 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI), pages 1–8. IEEE, 2024

    2024

  20. [28]

    U-sleep: resilient high-frequency sleep staging.NPJ digital medicine, 4(1):72, 2021

    MathiasPerslev, SuneDarkner, LykkeKempfner, MikiNikolic, PoulJørgenJennum, andChristian Igel. U-sleep: resilient high-frequency sleep staging.NPJ digital medicine, 4(1):72, 2021

    2021

  21. [29]

    The sleep heart health study: design, rationale, and methods.Sleep, 20(12):1077–1085, 1997

    Stuart F Quan, Barbara V Howard, Conrad Iber, James P Kiley, F Javier Nieto, George T O’Connor, David M Rapoport, Susan Redline, John Robbins, Jonathan M Samet, et al. The sleep heart health study: design, rationale, and methods.Sleep, 20(12):1077–1085, 1997

    1997

  22. [30]

    Language models are unsupervised multitask learners.OpenAI blog, 1(8):9, 2019

    Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. Language models are unsupervised multitask learners.OpenAI blog, 1(8):9, 2019. 13 OSF: On Pre-training and Scaling of Sleep Foundation Models

    2019

  23. [31]

    Sleep-disordered breathing in hispanic/latino individuals of diverse backgrounds

    Susan Redline, Daniela Sotres-Alvarez, Jose Loredo, Martica Hall, Sanjay R Patel, Alberto Ramos, Neomi Shah, Andrew Ries, Raanan Arens, Janice Barnhart, et al. Sleep-disordered breathing in hispanic/latino individuals of diverse backgrounds. the hispanic community health study...

    2014

  24. [32]

    The familial aggregation of obstructive sleep apnea

    Susan Redline, Peter V Tishler, Tor D Tosteson, John Williamson, Kenneth Kump, Ilene Browner, Veronica Ferrette, and Patrick Krejci. The familial aggregation of obstructive sleep apnea. American journal of respiratory and critical care medicine, 151(3):682–687, 1995

    1995

  25. [33]

    Prevalence and risk factors for sleep-disordered breathing in 8-to 11-year-old children: association with race and prematurity.The Journal of pediatrics, 142(4):383–389, 2003

    Carol L Rosen, Emma K Larkin, H Lester Kirchner, Judith L Emancipator, Sarah F Bivins, Susan A Surovec, Richard J Martin, and Susan Redline. Prevalence and risk factors for sleep-disordered breathing in 8-to 11-year-old children: association with race and prematurity.The Journ...

    2003

  26. [34]

    Quantifying the arousal threshold using polysomnography in obstructive sleep apnea.Sleep, 41(1):zsx183, 2018

    Scott A Sands, Philip I Terrill, Bradley A Edwards, Luigi Taranto Montemurro, Ali Azarbarzin, Melania Marques, Camila M De Melo, Stephen H Loring, James P Butler, David P White, et al. Quantifying the arousal threshold using polysomnography in obstructive sleep apnea.Sleep, 41...

    2018

  27. [35]

    Sleep-disordered breathing and cognition in older women.Journal of the American Geriatrics Society, 56(1):45–50, 2008

    Adam P Spira, Terri Blackwell, Katie L Stone, Susan Redline, Jane A Cauley, Sonia Ancoli-Israel, and Kristine Yaffe. Sleep-disordered breathing and cognition in older women.Journal of the American Geriatrics Society, 56(1):45–50, 2008

    2008

  28. [36]

    Sleepfm: Multi-modal representation learning for sleep across brain activity, ecg and respiratory signals.arXiv preprint arXiv:2405.17766, 2024

    Rahul Thapa, Bryan He, Magnus Ruud Kjaer, Hyatt Moore, Gauri Ganjoo, Emmanuel Mignot, and James Zou. Sleepfm: Multi-modal representation learning for sleep across brain activity, ecg and respiratory signals.arXiv preprint arXiv:2405.17766, 2024

    2024 arXiv

  29. [37]

    A multimodal sleep foundation model for disease prediction.Nature Medicine, pages 1–11, 2026

    Rahul Thapa, Magnus Ruud Kjaer, Bryan He, Ian Covert, Hyatt Moore IV, Umaer Hanif, Gauri Ganjoo, M Brandon Westover, Poul Jennum, Andreas Brink-Kjaer, et al. A multimodal sleep foundation model for disease prediction.Nature Medicine, pages 1–11, 2026

    2026

  30. [38]

    Neural discrete representation learning.Advances in neural information processing systems, 30, 2017

    Aaron Van Den Oord, Oriol Vinyals, et al. Neural discrete representation learning.Advances in neural information processing systems, 30, 2017

    2017

  31. [39]

    Simper: Simple self-supervised learning of periodic targets

    Yuzhe Yang, Xin Liu, Jiang Wu, Silviu Borac, Dina Katabi, Ming-Zher Poh, and Daniel McDuff. Simper: Simple self-supervised learning of periodic targets. InThe Eleventh International Conference on Learning Representations, 2023

    2023

  32. [40]

    Artificial intelligence-enabled detection and assessment of parkinson’s disease using nocturnal breathing signals.Nature Medicine, 28(10):2207–2215, 2022

    Yuzhe Yang, Yuan Yuan, Guo Zhang, Hao Wang, Ying-Cong Chen, Yingcheng Liu, Christopher G Tarolli, Daniel Crepeau, Jan Bukartyk, Mithri R Junna, et al. Artificial intelligence-enabled detection and assessment of parkinson’s disease using nocturnal breathing signals.Nature Medic...

    2022

  33. [41]

    Burden of sleep apnea: rationale, design, and major findings of the wisconsin sleep cohort study

    Terry Young, Mari Palta, Jerome Dempsey, Paul E Peppard, F Javier Nieto, and K Mae Hla. Burden of sleep apnea: rationale, design, and major findings of the wisconsin sleep cohort study. WMJ: official publication of the State Medical Society of Wisconsin, 108(5):246, 2009

    2009

  34. [42]

    Self-supervised learning for human activity recognition using 700,000 person-days of wearable data.NPJ digital medicine, 7(1):91, 2024

    Hang Yuan, Shing Chan, Andrew P Creagh, Catherine Tong, Aidan Acquah, David A Clifton, and Aiden Doherty. Self-supervised learning for human activity recognition using 700,000 person-days of wearable data.NPJ digital medicine, 7(1):91, 2024

    2024

  35. [43]

    Challenges and opportunities of deep learning for wearable-based objective sleep assessment.npj Digital Medicine, 7(1):85, 2024

    Bing Zhai, Greg J Elder, and Alan Godfrey. Challenges and opportunities of deep learning for wearable-based objective sleep assessment.npj Digital Medicine, 7(1):85, 2024. 14 OSF: On Pre-training and Scaling of Sleep Foundation Models

    2024

  36. [44]

    Thenationalsleepresearchresource: towardsa sleep data commons.Journal of the American Medical Informatics Association, 25(10):1351–1358, 2018

    Guo-Qiang Zhang, Licong Cui, Remo Mueller, Shiqiang Tao, Matthew Kim, Michael Rueschman, SaraMariani, DanielMobley, andSusanRedline. Thenationalsleepresearchresource: towardsa sleep data commons.Journal of the American Medical Informatics Association, 25(10):1351–1358, 2018

    2018

  37. [45]

    Sensorlm: Learning the language of wearable sensors.arXiv preprint arXiv:2506.09108, 2025

    Yuwei Zhang, Kumar Ayush, Siyuan Qiao, A Ali Heydari, Girish Narayanswamy, Maxwell A Xu, Ahmed A Metwally, Shawn Xu, Jake Garrison, Xuhai Xu, et al. Sensorlm: Learning the language of wearable sensors.arXiv preprint arXiv:2506.09108, 2025. 15 OSF: On Pre-training and Scaling o...

    2025 arXiv

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.