mlsanityData to 7 Oct 2026

Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions

arXiv 2507.00191 ↗First posted 30 Jun 2025cs.LG cs.AIcited by 18

Not on the map. Why?

Abstract (arXiv)

Wearable devices record physiological and behavioral signals that can improve health predictions. While foundation models are increasingly used for such predictions, they have been primarily applied to low-level sensor data, despite behavioral data often being more informative due to their alignment with physiologically relevant timescales and quantities. We develop foundation models of such behavioral signals using over 2.5B hours of wearable data from 162K individuals, systematically optimizing architectures and tokenization strategies for this unique dataset. Evaluated on 57 health-related tasks, our model shows strong performance across diverse real-world applications including individual-level classification and time-varying health state prediction. The model excels in behavior-driven tasks like sleep prediction, and improves further when combined with representations of raw sensor data. These results underscore the importance of tailoring foundation model design to wearables and demonstrate the potential to enable new health applications.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 33 of its 54 references that are in the map. First posted on arXiv 30 Jun 2025.

Builds on

33 of 54 references.

References · 33

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  3. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  4. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  5. 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
  6. 2019Root Mean Square Layer Normalizationcited by 1,425
  7. 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,283
  8. 2018Self-Attention with Relative Position Representationscited by 1,112
  9. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 982
  10. 2024Chronos: Learning the Language of Time Seriescited by 712
  11. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
  12. 2020A Transformer-based Framework for Multivariate Time Series Representation Learningcited by 449
21 more
  1. 2021Pay Attention to MLPscited by 367
  2. 2021Multi-Time Attention Networks for Irregularly Sampled Time Seriescited by 234
  3. 2023TimeGPT-1cited by 149
  4. 2019MIMIC-Extract: A Data Extraction, Preprocessing, and Representation Pipeline for MIMIC-IIIcited by 121
  5. 2018Spreading vectors for similarity searchcited by 114
  6. 2024Is Mamba Effective for Time Series Forecasting?cited by 110
  7. 2021Position Information in Transformers: An Overviewcited by 103
  8. 2021Self-Supervised Transformer for Sparse and Irregularly Sampled Multivariate Clinical Time-Seriescited by 96
  9. 2023Large-scale Training of Foundation Models for Wearable Biosignalscited by 89
  10. 2016Modeling Missing Data in Clinical Time Series with RNNscited by 80
  11. 2022Self-supervised Learning for Human Activity Recognition Using 700,000 Person-days of Wearable Datacited by 71
  12. 2024Scaling Wearable Foundation Modelscited by 57
  13. 2023Warpformer: A Multi-scale Modeling Approach for Irregular Clinical Time Seriescited by 54
  14. 2018RAIM: Recurrent Attentive and Intensive Model of Multimodal Patient Monitoring Datacited by 47
  15. 2024Context Clues: Evaluating Long Context Models for Clinical Prediction Tasks on EHRscited by 39
  16. 2024EHRMamba: Towards Generalizable and Scalable Foundation Models for Electronic Health Recordscited by 31
  17. 2023DuETT: Dual Event Time Transformer for Electronic Health Recordscited by 17
  18. 2024RelCon: Relative Contrastive Learning for a Motion Foundation Model for Wearable Datacited by 17
  19. 2024Wearable Accelerometer Foundation Models for Health via Knowledge Distillationcited by 15
  20. 2023Event-Based Contrastive Learning for Medical Time Seriescited by 13
  21. 2022Self-supervised Pretraining and Transfer Learning Enable Flu and COVID-19 Predictions in Small Mobile Sensing Datasetscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 7; 2026: 11.

18 papers.

Sort:

By area:

  1. 2025HiMAE: Hierarchical Masked Autoencoders Discover Resolution-Specific Structure in Wearable Time Seriescited by 19
  2. 2025CLEF: Clinically-Guided Contrastive Learning for Electrocardiogram Foundation Modelscited by 7
  3. 2025PPG-Distill: Efficient Photoplethysmography Signals Analysis via Foundation Model Distillationcited by 6
  4. 2026Wavelet-Driven Masked Multiscale Reconstruction for PPG Foundation Modelscited by 4
  5. 2026Seg-MoE: Multi-Resolution Segment-wise Mixture-of-Experts for Time Series Forecasting Transformerscited by 3
  6. 2026Wearable Foundation Models Should Go Beyond Static Encoderscited by 3
  7. 2026Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learningcited by 3
  8. 2025SlotFM: A Motion Foundation Model with Slot Attention for Diverse Downstream Taskscited by 1
  9. 2025Speech Foundation Models Generalize to Time Series Tasks from Wearable Sensor Datacited by 0
  10. 2025Generalist vs Specialist Time Series Foundation Models: Investigating Potential Emergent Behaviors in Assessing Human Health Using PPG Signalscited by 0
  11. 2025Wrist Photoplethysmography Predicts Dietary Informationcited by 0
  12. 2026MoHETS: Long-term Time Series Forecasting with Mixture-of-Heterogeneous-Expertscited by 0
6 more
  1. 2026PerCaM-Health: Personalized Dynamic Causal Graphs for Healthcare Reasoningcited by 0
  2. 2026WavesFM: Hierarchical Representation Learning for Longitudinal Wearable Sensor Waveformscited by 0
  3. 2026VCR: Learning Valid Contextual Representation for Incomplete Wearable Signalscited by 0
  4. 2026Retrieval-Augmented Personalization with Foundation Models for Wearable Stress Detectioncited by 0
  5. 2026Physical activities enable scalable foundation modelling for broad-spectrum health predictioncited by 0
  6. 2026Foundation models for movement data: Are they ready for prime-time?cited by 0