pith. sign in

arxiv: 2606.18147 · v1 · pith:IIP6ZF72new · submitted 2026-06-16 · 💻 cs.AI

WEQA: Wearable hEalth Question Answering with Query-Adaptive Agentic Reasoning

Yuwei Zhang , Tong Xia , Bianca Emmerich , Yu Yvonne Wu , Dimitris Spathis , Xin Liu , Daniel McDuff , Cecilia Mascolo This is my paper

Pith reviewed 2026-06-27 00:50 UTC · model grok-4.3

classification 💻 cs.AI
keywords wearable health dataquestion answeringagentic reasoningLLM controllersensor analysismedical QAadaptive frameworkbenchmark datasets
0
0 comments X

The pith

WEQA routes wearable health queries adaptively via an LLM controller to sensor tools and models for higher accuracy.

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

Wearable devices produce continuous high-dimensional longitudinal data that resists direct use by text-trained language models. Fixed workflows cannot adapt to the range of sensor modalities and user intents in health questions. The paper introduces an agent framework in which an LLM controller creates execution plans, routes each query to the right mix of analysis tools and pretrained models, and audits the output against external knowledge. This yields more accurate answers on a benchmark of four open datasets spanning analytic and predictive tasks in three health domains, plus higher usefulness ratings from medical experts.

Core claim

The central claim is that an LLM controller synthesizing execution plans, dynamically routing queries to combinations of sensor analysis and pretrained models, and performing grounded auditing enables accurate answers to questions about continuous wearable health data, where fixed reasoning workflows or single foundation models fall short.

What carries the argument

The LLM controller that synthesizes execution plans, routes queries to sensor tools and models, and audits responses with external knowledge.

If this is right

  • The approach handles diversity of sensor modalities and user intents that defeat fixed workflows.
  • Accuracy improves by 24 percent relative to LLM and agentic baselines on the curated benchmark.
  • Blinded evaluation by medical experts records gains in usefulness and clinical soundness.
  • The framework applies across analytic and predictive tasks drawn from multiple open wearable datasets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same routing logic could be tested on live streaming sensor data for real-time health alerts.
  • Similar controllers might address other longitudinal multimodal sources such as continuous glucose or sleep records.
  • Grounded auditing steps could reduce the rate of unsupported health recommendations in consumer apps.

Load-bearing premise

An LLM can reliably synthesize correct execution plans, route queries without errors, and audit outputs without introducing hallucinations.

What would settle it

A set of held-out wearable queries where the controller selects incorrect tools or produces audited answers that contradict the underlying sensor data.

Figures

Figures reproduced from arXiv: 2606.18147 by Bianca Emmerich, Cecilia Mascolo, Daniel McDuff, Dimitris Spathis, Tong Xia, Xin Liu, Yuwei Zhang, Yu Yvonne Wu.

Figure 1
Figure 1. Figure 1: Examples of heterogeneous wearable health questions and the computational requirements. The right panel denotes representative computation steps, such as temporal analysis and model prediction. conversational health assistance increasingly plau￾sible (OpenAI, 2026). Language models can be used to interpret free-form questions, provide ex￾planations and reasoning, and draw on broad med￾ical knowledge acquir… view at source ↗
Figure 2
Figure 2. Figure 2: WEQA query-adaptive agent framework for wearable health question answering. Given a user query, wearable data, and optional personal history, the agent plans a query-specific workflow, gathers evidence through sensor analysis and adaptive prediction, and performs grounded safety auditing producing a safety-aware response. 2024; Cosentino et al., 2024). While simple and flexible, text-based representations … view at source ↗
Figure 3
Figure 3. Figure 3: Examples of query-adaptive reasoning pathways in WEQA for heterogeneous wearable health questions. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Overview of the proposed wearable health QA benchmark. (a) Four task families with representative example questions. (b) Fine-grained subtask taxonomy within each family. diction as definitive evidence. Personalization is handled according to the availability of user history. When self-labeled personal history are available, the agent performs lightweight few-shot adaptation of the population-level model o… view at source ↗
Figure 5
Figure 5. Figure 5: Efficiency analysis across wearable health QA systems. WEQA achieves lower latency and re￾duced overhead compared to prior agent baselines. 0 2 4 Correctness/ Grounding Personalization Usefulness Clinical Plausibility Experts 0 2 4 Users WEQA React Multi-Agent [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Human evaluation Results. Medical expert (left) and user (right) ratings across four dimensions for three models; higher scores indicate better performance. agent orchestration (72.0 /157.2). Text-only and multimodal prompting perform poorly, indicating that even short-horizon numeric questions require explicit sensor-grounded computation. The advan￾tage becomes larger on Long-Horizon Analytical QA, which … view at source ↗
read the original abstract

Language models are remarkably capable at medical question answering, in some cases surpassing the accuracy of general physicians. However, answering questions about wearable health data remains challenging and understudied, as these ubiquitous sensors produce continuous, high-dimensional, and longitudinal data, which is non-trivial to align with text-centric distributions in LLM pretraining. The diversity of sensor modalities and user intents cannot be effectively handled by a fixed reasoning workflow or a single pretrained foundation model. To address these challenges, we propose WEQA, a query-adaptive agent framework that unifies LLM reasoning with specialized wearable analytical and modeling tools. An LLM controller is employed to synthesize execution plans and dynamically route each query to the appropriate combination of sensor analysis and pretrained models, and perform grounded response auditing with external knowledge. We also curate a benchmark spanning four open wearable datasets comprising analytic and predictive tasks in three different health domains. Experiments show that our framework is 24% more accurate than LLM and agentic baselines, and a blinded study with 12 medical experts and 8 users shows substantial gains in usefulness and clinical soundness.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 1 minor

Summary. The manuscript proposes WEQA, a query-adaptive agent framework for wearable health question answering that unifies LLM reasoning with specialized sensor analysis and modeling tools. An LLM controller synthesizes execution plans, dynamically routes queries to appropriate tools or models based on modality and intent, and performs grounded auditing with external knowledge. The authors curate a benchmark from four open wearable datasets spanning analytic and predictive tasks across three health domains. Experiments are reported to show a 24% accuracy improvement over LLM and agentic baselines, with a blinded study of 12 medical experts and 8 users indicating gains in usefulness and clinical soundness.

Significance. If the empirical claims hold and the LLM controller operates reliably, the work would advance wearable health QA by demonstrating that adaptive, query-dependent routing can address the limitations of fixed workflows for high-dimensional, longitudinal sensor data. The curation of a multi-dataset benchmark across domains is a concrete contribution that could support future reproducible comparisons.

major comments (3)
  1. [Abstract] Abstract: The central claim of a 24% accuracy gain over baselines is load-bearing for the paper's contribution, yet the abstract (and provided text) supplies no baseline definitions, dataset splits, statistical tests, or error analysis, preventing verification that the data supports the result.
  2. [Abstract] Abstract: The superiority over fixed workflows rests on the assumption that the LLM controller reliably synthesizes plans, routes queries without misrouting, and audits without hallucinations; however, no metrics on controller error rates, plan correctness, or failure cases are reported to substantiate this.
  3. [Abstract] Abstract: The blinded expert/user study claims substantial gains in usefulness and clinical soundness, but lacks any description of study design, quantitative metrics, or statistical evaluation, making the clinical claim impossible to assess.
minor comments (1)
  1. [Abstract] The abstract uses stylized capitalization in the title acronym (hEalth) without explanation; consider standardizing for clarity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and will revise the manuscript to improve clarity and verifiability of the empirical claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim of a 24% accuracy gain over baselines is load-bearing for the paper's contribution, yet the abstract (and provided text) supplies no baseline definitions, dataset splits, statistical tests, or error analysis, preventing verification that the data supports the result.

    Authors: The full manuscript defines baselines (LLM-only, fixed-workflow agents, and single-model approaches) in Section 4.2, details the four-dataset curation and train/test splits in Section 3.2, reports paired statistical tests with p-values in Section 4.3, and includes error analysis with per-task breakdowns in Section 4.4. We will revise the abstract to briefly reference these elements and note that gains are statistically supported, improving standalone verifiability without altering the reported 24% figure. revision: yes

  2. Referee: [Abstract] Abstract: The superiority over fixed workflows rests on the assumption that the LLM controller reliably synthesizes plans, routes queries without misrouting, and audits without hallucinations; however, no metrics on controller error rates, plan correctness, or failure cases are reported to substantiate this.

    Authors: We agree that controller-specific metrics would strengthen the claims. The current evaluation emphasizes end-to-end task accuracy, but we will add a dedicated analysis in the revision quantifying plan correctness (via manual review of 200 queries), routing accuracy, and categorized failure cases, to be presented in a new subsection of Section 4. revision: yes

  3. Referee: [Abstract] Abstract: The blinded expert/user study claims substantial gains in usefulness and clinical soundness, but lacks any description of study design, quantitative metrics, or statistical evaluation, making the clinical claim impossible to assess.

    Authors: Section 5 of the manuscript fully describes the blinded protocol, participant recruitment (12 experts, 8 users), 5-point Likert scales for usefulness and soundness, and statistical comparisons (Wilcoxon signed-rank tests). We will expand the abstract to concisely summarize the study design and report the key quantitative gains and significance levels. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical framework evaluation with no derivations or self-referential reductions

full rationale

The paper describes an agentic LLM-based framework for wearable health QA, justifies the design by noting limitations of fixed workflows, and reports empirical accuracy gains plus expert/user studies. No equations, fitted parameters, predictions derived from inputs, or self-citation chains appear in the abstract or described content. The central claims rest on experimental comparisons and blinded evaluations that are independent of any definitional or fitted-input circularity. The derivation chain consists of system architecture choices motivated by stated challenges, not reductions to the paper's own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no details on free parameters, axioms, or invented entities; assessment is limited to the high-level description of the framework and benchmark.

pith-pipeline@v0.9.1-grok · 5741 in / 1189 out tokens · 46979 ms · 2026-06-27T00:50:09.345373+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

217 extracted references · 1 canonical work pages

  1. [1]

    Scientific Data , volume=

    TILES-2018, a longitudinal physiologic and behavioral data set of hospital workers , author=. Scientific Data , volume=. 2020 , publisher=

    2018

  2. [2]

    Advances in Neural Information Processing Systems , volume=

    Towards open respiratory acoustic foundation models: Pretraining and benchmarking , author=. Advances in Neural Information Processing Systems , volume=

  3. [3]

    Scientific data , volume=

    A new, short-recorded photoplethysmogram dataset for blood pressure monitoring in China , author=. Scientific data , volume=. 2018 , publisher=

    2018

  4. [4]

    Aho and Jeffrey D

    Alfred V. Aho and Jeffrey D. Ullman , title =. 1972

    1972

  5. [5]

    Publications Manual , year = "1983", publisher =

    1983

  6. [6]

    Chandra and Dexter C

    Ashok K. Chandra and Dexter C. Kozen and Larry J. Stockmeyer , year = "1981", title =. doi:10.1145/322234.322243

    work page doi:10.1145/322234.322243 1981

  7. [7]

    Scalable training of

    Andrew, Galen and Gao, Jianfeng , booktitle=. Scalable training of

  8. [8]

    Dan Gusfield , title =. 1997

    1997

  9. [9]

    Tetreault , title =

    Mohammad Sadegh Rasooli and Joel R. Tetreault , title =. Computing Research Repository , volume =. 2015 , url =

    2015

  10. [10]

    A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data , Volume =

    Ando, Rie Kubota and Zhang, Tong , Issn =. A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data , Volume =. Journal of Machine Learning Research , Month = dec, Numpages =

  11. [11]

    Introducing ChatGPT Health , year =

  12. [12]

    Advancing Claude in healthcare and the life sciences , year =

  13. [13]

    Nature communications , volume=

    GWAS identifies 14 loci for device-measured physical activity and sleep duration , author=. Nature communications , volume=. 2018 , publisher=

    2018

  14. [14]

    arXiv preprint arXiv:2507.06261 , year=

    Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities , author=. arXiv preprint arXiv:2507.06261 , year=

    Pith/arXiv arXiv

  15. [15]

    Nature medicine , volume=

    Wearable-device-measured physical activity and future health risk , author=. Nature medicine , volume=. 2020 , publisher=

    2020

  16. [16]

    Conference on Neural Information Processing Systems (NeurIPS) , year=

    SensorLM: Learning the Language of Wearable Sensors , author=. Conference on Neural Information Processing Systems (NeurIPS) , year=

  17. [17]

    Proceedings of the IEEE/CVF international conference on computer vision , pages=

    Sigmoid loss for language image pre-training , author=. Proceedings of the IEEE/CVF international conference on computer vision , pages=

  18. [18]

    arXiv preprint arXiv:2308.13703 , year=

    PAITS: pretraining and augmentation for irregularly-sampled time series , author=. arXiv preprint arXiv:2308.13703 , year=

    arXiv

  19. [19]

    Forty-second International Conference on Machine Learning , year=

    Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions , author=. Forty-second International Conference on Machine Learning , year=

  20. [20]

    The Thirteenth International Conference on Learning Representations , year=

    Scaling Wearable Foundation Models , author=. The Thirteenth International Conference on Learning Representations , year=

  21. [21]

    European Conference on Computer Vision (ECCV) , year=

    On Multi-Domain Long-Tailed Recognition, Imbalanced Domain Generalization and Beyond , author=. European Conference on Computer Vision (ECCV) , year=

  22. [22]

    NeurIPS , year=

    Neighbourhood components analysis , author=. NeurIPS , year=

  23. [23]

    International Conference on Machine Learning , pages=

    Delving into deep imbalanced regression , author=. International Conference on Machine Learning , pages=. 2021 , organization=

    2021

  24. [24]

    International Conference on Learning Representations , year=

    SimPer: Simple Self-Supervised Learning of Periodic Targets , author=. International Conference on Learning Representations , year=

  25. [25]

    International conference on machine learning , pages=

    A simple framework for contrastive learning of visual representations , author=. International conference on machine learning , pages=. 2020 , organization=

    2020

  26. [26]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Targeted supervised contrastive learning for long-tailed recognition , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  27. [27]

    Advances in Neural Information Processing Systems , volume=

    Supervised contrastive learning , author=. Advances in Neural Information Processing Systems , volume=

  28. [28]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Momentum contrast for unsupervised visual representation learning , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  29. [29]

    arXiv preprint arXiv:2003.04297 , year=

    Improved baselines with momentum contrastive learning , author=. arXiv preprint arXiv:2003.04297 , year=

    Pith/arXiv arXiv 2003

  30. [30]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    Deep residual learning for image recognition , author=. Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

  31. [31]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Emerging properties in self-supervised vision transformers , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

  32. [32]

    Scaling Learning Algorithms Towards

    Bengio, Yoshua and LeCun, Yann , booktitle =. Scaling Learning Algorithms Towards

  33. [33]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Scaling vision transformers , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  34. [34]

    Nature Medicine , volume=

    Artificial intelligence-enabled detection and assessment of Parkinson's disease using nocturnal breathing signals , author=. Nature Medicine , volume=. 2022 , publisher=

    2022

  35. [35]

    International Conference on Learning Representations , year=

    Decoupled Weight Decay Regularization , author=. International Conference on Learning Representations , year=

  36. [36]

    International Conference on Learning Representations , year=

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale , author=. International Conference on Learning Representations , year=

  37. [37]

    Journal of machine learning research , volume=

    Exploring the limits of transfer learning with a unified text-to-text transformer , author=. Journal of machine learning research , volume=

  38. [38]

    arXiv preprint arXiv:1712.00409 , year=

    Deep learning scaling is predictable, empirically , author=. arXiv preprint arXiv:1712.00409 , year=

    Pith/arXiv arXiv

  39. [39]

    arXiv preprint arXiv:2305.15525 , year=

    Large language models are few-shot health learners , author=. arXiv preprint arXiv:2305.15525 , year=

    arXiv

  40. [40]

    arXiv preprint arXiv:2010.14701 , year=

    Scaling laws for autoregressive generative modeling , author=. arXiv preprint arXiv:2010.14701 , year=

    Pith/arXiv arXiv 2010

  41. [41]

    International Conference on Machine Learning , pages=

    Scaling laws for generative mixed-modal language models , author=. International Conference on Machine Learning , pages=. 2023 , organization=

    2023

  42. [42]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    On data scaling in masked image modeling , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

  43. [43]

    and Osindero, Simon and Teh, Yee Whye , journal =

    Hinton, Geoffrey E. and Osindero, Simon and Teh, Yee Whye , journal =. A Fast Learning Algorithm for Deep Belief Nets , volume =

  44. [44]

    arXiv preprint arXiv:2402.03714 , year=

    Advancing Location-Invariant and Device-Agnostic Motion Activity Recognition on Wearable Devices , author=. arXiv preprint arXiv:2402.03714 , year=

    arXiv

  45. [45]

    2016 , publisher=

    Deep learning , author=. 2016 , publisher=

    2016

  46. [46]

    NPJ digital medicine , volume=

    Self-supervised learning for human activity recognition using 700,000 person-days of wearable data , author=. NPJ digital medicine , volume=. 2024 , publisher=

    2024

  47. [47]

    arXiv preprint arXiv:2408.08849 , year=

    ECG-Chat: A Large ECG-Language Model for Cardiac Disease Diagnosis , author=. arXiv preprint arXiv:2408.08849 , year=

    arXiv

  48. [48]

    Rahul Thapa and Bryan He and Magnus Ruud Kjaer and Hyatt Moore IV and Gauri Ganjoo and Emmanuel Mignot and James Zou , booktitle=. Sleep. 2024 , url=

    2024

  49. [49]

    The Twelfth International Conference on Learning Representations , year=

    Large-scale training of foundation models for wearable biosignals , author=. The Twelfth International Conference on Learning Representations , year=

  50. [50]

    Forty-first International Conference on Machine Learning , year=

    A decoder-only foundation model for time-series forecasting , author=. Forty-first International Conference on Machine Learning , year=

  51. [51]

    2024 , url=

    Mononito Goswami and Konrad Szafer and Arjun Choudhry and Yifu Cai and Shuo Li and Artur Dubrawski , booktitle=. 2024 , url=

    2024

  52. [52]

    IEEE communications surveys & tutorials , volume=

    Federated learning in mobile edge networks: A comprehensive survey , author=. IEEE communications surveys & tutorials , volume=. 2020 , publisher=

    2020

  53. [53]

    Transactions on Machine Learning Research , issn=

    Chronos: Learning the Language of Time Series , author=. Transactions on Machine Learning Research , issn=. 2024 , url=

    2024

  54. [54]

    arXiv preprint arXiv:2409.00088 , year=

    On-device language models: A comprehensive review , author=. arXiv preprint arXiv:2409.00088 , year=

    arXiv

  55. [55]

    arXiv preprint arXiv:2310.03589 , year=

    TimeGPT-1 , author=. arXiv preprint arXiv:2310.03589 , year=

    arXiv

  56. [56]

    Lag-llama: Towards foundation models for time series forecasting , author=

  57. [57]

    Journal of medical Internet research , volume=

    Fitbit-based interventions for healthy lifestyle outcomes: systematic review and meta-analysis , author=. Journal of medical Internet research , volume=. 2020 , publisher=

    2020

  58. [58]

    Circulation , volume=

    Detection of atrial fibrillation in a large population using wearable devices: the Fitbit Heart Study , author=. Circulation , volume=. 2022 , publisher=

    2022

  59. [59]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Masked autoencoders are scalable vision learners , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

  60. [60]

    Annals of the New York Academy of Sciences , volume=

    Mobile health: the power of wearables, sensors, and apps to transform clinical trials , author=. Annals of the New York Academy of Sciences , volume=. 2016 , publisher=

    2016

  61. [61]

    Advances in Neural Information Processing Systems , volume=

    Masked autoencoders that listen , author=. Advances in Neural Information Processing Systems , volume=

  62. [62]

    Advances in neural information processing systems , volume=

    Discriminative unsupervised feature learning with convolutional neural networks , author=. Advances in neural information processing systems , volume=

  63. [63]

    European conference on computer vision , pages=

    Unsupervised learning of visual representations by solving jigsaw puzzles , author=. European conference on computer vision , pages=. 2016 , organization=

    2016

  64. [64]

    Proceedings of the European conference on computer vision (ECCV) , pages=

    Deep clustering for unsupervised learning of visual features , author=. Proceedings of the European conference on computer vision (ECCV) , pages=

  65. [65]

    International Conference on Learning Representations , year=

    Unsupervised Representation Learning by Predicting Image Rotations , author=. International Conference on Learning Representations , year=

  66. [66]

    International Conference on Learning Representations (ICLR) , year=

    Unsupervised representation learning by predicting image rotations , author=. International Conference on Learning Representations (ICLR) , year=

  67. [67]

    Proceedings of the IEEE international conference on computer vision , pages=

    Representation learning by learning to count , author=. Proceedings of the IEEE international conference on computer vision , pages=

  68. [68]

    The Eleventh International Conference on Learning Representations , year=

    SimPer: Simple Self-Supervised Learning of Periodic Targets , author=. The Eleventh International Conference on Learning Representations , year=

  69. [69]

    bioRxiv , pages=

    What Does Large-scale Electrodermal Sensing Reveal? , author=. bioRxiv , pages=. 2024 , publisher=

    2024

  70. [70]

    Proceedings of the National Academy of Sciences , volume=

    Explaining neural scaling laws , author=. Proceedings of the National Academy of Sciences , volume=. 2024 , publisher=

    2024

  71. [71]

    arXiv preprint arXiv:2001.08361 , year=

    Scaling laws for neural language models , author=. arXiv preprint arXiv:2001.08361 , year=

    Pith/arXiv arXiv 2001

  72. [72]

    JMIR formative research , volume=

    Heart rate measurement accuracy of Fitbit Charge 4 and Samsung Galaxy Watch Active2: device evaluation study , author=. JMIR formative research , volume=. 2022 , publisher=

    2022

  73. [73]

    Frontiers in public health , volume=

    An overview of heart rate variability metrics and norms , author=. Frontiers in public health , volume=. 2017 , publisher=

    2017

  74. [74]

    The Lancet Digital Health , volume=

    Heart rate variability with photoplethysmography in 8 million individuals: a cross-sectional study , author=. The Lancet Digital Health , volume=. 2020 , publisher=

    2020

  75. [75]

    Conference on Health, Inference, and Learning , pages=

    Self-supervised pretraining and transfer learning enable titlebreak flu and covid-19 predictions in small mobile sensing datasets , author=. Conference on Health, Inference, and Learning , pages=. 2023 , organization=

    2023

  76. [76]

    arXiv preprint arXiv:2307.09288 , year=

    Llama 2: Open foundation and fine-tuned chat models , author=. arXiv preprint arXiv:2307.09288 , year=

    Pith/arXiv arXiv

  77. [77]

    Merrill and Mingtian Tan and Vinayak Gupta and Thomas Hartvigsen and Tim Althoff , title=

    Mike A. Merrill and Mingtian Tan and Vinayak Gupta and Thomas Hartvigsen and Tim Althoff , title=. 2024 , cdate=

    2024

  78. [78]

    European Conference on Computer Vision , pages=

    Masked siamese networks for label-efficient learning , author=. European Conference on Computer Vision , pages=. 2022 , organization=

    2022

  79. [79]

    Conference on Health, Inference, and Learning , pages=

    Homekit2020: A benchmark for time series classification on a large mobile sensing dataset with laboratory tested ground truth of influenza infections , author=. Conference on Health, Inference, and Learning , pages=. 2023 , organization=

    2023

  80. [80]

    Proceedings of Neural Information Processing Systems(NeurIPS) , month =

    Balanced Meta-Softmax for Long-Tailed Visual Recognition , author=. Proceedings of Neural Information Processing Systems(NeurIPS) , month =

Showing first 80 references.