mlsanityData to 14 Oct 2026

Wearable Foundation Models Should Go Beyond Static Encoders

arXiv 2603.19564 ↗First posted 20 Mar 2026cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Wearable foundation models (WFMs), trained on large volumes of data collected by affordable, always-on devices, have demonstrated strong performance on short-term, well-defined health monitoring tasks, including activity recognition, fitness tracking, and cardiovascular signal assessment. However, most existing WFMs primarily map short temporal windows to predefined labels via static encoders, emphasizing retrospective prediction rather than reasoning over evolving personal history, context, and future risk trajectories. As a result, they are poorly suited for modeling chronic, progressive, or episodic health conditions that unfold over weeks, months or years. Hence, we argue that WFMs must move beyond static encoders and be explicitly designed for longitudinal, anticipatory health reasoning. We identify three foundational shifts required to enable this transition: (1) Structurally rich data, which goes beyond isolated datasets or outcome-conditioned collection to integrated multimodal, long-term personal trajectories, and contextual metadata, ideally supported by open and interoperable data ecosystems; (2) Longitudinal-aware multimodal modeling, which prioritizes long-context inference, temporal abstraction, and personalization over cross-sectional or population-level prediction; and (3) Agentic inference systems, which move beyond static prediction to support planning, decision-making, and clinically grounded intervention under uncertainty. Together, these shifts reframe wearable health monitoring from retrospective signal interpretation toward continuous, anticipatory, and human-aligned health support.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 31 of its 68 references that are in the map. First posted on arXiv 20 Mar 2026.

Builds on

31 of 68 references.

References · 31

  1. 2017Attention Is All You Needcited by 60,361
  2. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,718
  3. 2024Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCIcited by 327
  4. 2023Auto-GPT for Online Decision Making: Benchmarks and Additional Opinionscited by 255
  5. 2023Large Language Models are Few-Shot Health Learnerscited by 118
  6. 2023Large-scale Training of Foundation Models for Wearable Biosignalscited by 95
  7. 2024PaPaGei: Open Foundation Models for Optical Physiological Signalscited by 69
  8. 2024Health-LLM: Large Language Models for Health Prediction via Wearable Sensor Datacited by 66
  9. 2024Scaling Wearable Foundation Modelscited by 63
  10. 2025SensorLM: Learning the Language of Wearable Sensorscited by 49
  11. 2025ContextAgent: Context-Aware Proactive LLM Agents with Open-World Sensory Perceptionscited by 45
  12. 2024SleepFM: Multi-modal Representation Learning for Sleep Across Brain Activity, ECG and Respiratory Signalscited by 41
19 more
  1. 2024SensorLLM: Aligning Large Language Models with Motion Sensors for Human Activity Recognitioncited by 38
  2. 2025OpenTSLM: Time-Series Language Models for Reasoning over Multivariate Medical Text- and Time-Series Datacited by 36
  3. 2021Intra-Inter Subject Self-supervised Learning for Multivariate Cardiac Signalscited by 35
  4. 2024Toward Foundation Model for Multivariate Wearable Sensing of Physiological Signalscited by 32
  5. 2024MEIT: Multimodal Electrocardiogram Instruction Tuning on Large Language Models for Report Generationcited by 29
  6. 2022GLOBEM Dataset: Multi-Year Datasets for Longitudinal Human Behavior Modeling Generalizationcited by 28
  7. 2025BrainOmni: A Brain Foundation Model for Unified EEG and MEG Signalscited by 27
  8. 2023Frequency-Aware Masked Autoencoders for Multimodal Pretraining on Biosignalscited by 23
  9. 2025LSM-2: Learning from Incomplete Wearable Sensor Datacited by 23
  10. 2024Transforming Wearable Data into Personal Health Insights using Large Language Model Agentscited by 19
  11. 2025Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictionscited by 18
  12. 2023Sequential Multi-Dimensional Self-Supervised Learning for Clinical Time Seriescited by 16
  13. 2024MedTsLLM: Leveraging LLMs for Multimodal Medical Time Series Analysiscited by 15
  14. 2024Promoting cross-modal representations to improve multimodal foundation models for physiological signalscited by 13
  15. 2025PhysioWave: A Multi-Scale Wavelet-Transformer for Physiological Signal Representationcited by 12
  16. 2025Towards Robust Multimodal Physiological Foundation Models: Handling Arbitrary Missing Modalitiescited by 11
  17. 2024RespLLM: Unifying Audio and Text with Multimodal LLMs for Generalized Respiratory Health Predictioncited by 10
  18. 2021MAUS: A Dataset for Mental Workload Assessmenton N-back Task Using Wearable Sensorcited by 7
  19. 2025GLOSS: Group of LLMs for Open-Ended Sensemaking of Passive Sensing Data for Health and Wellbeingcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026WavesFM: Hierarchical Representation Learning for Longitudinal Wearable Sensor Waveformscited by 0
  2. 2026Longitudinal Multimodal Sensing of Physical Activity and Well-Being in Older Adultscited by 0
  3. 2026BALMS: Benchmarking Agentic LLMs for Longitudinal Mental Health Sensingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Towards a General Intelligence and Interface for Wearable Health Data
  2. 2026HealthCAT: An Interpretable Encoder-only Transformer Framework for Health Indicator Prediction and Temporal Interpretation of Wearable Sensor Data
  3. 2026WearableQA: A Benchmark for Health Reasoning over Real-World Wearable Data
  4. 2026FemWear: A Parameter-Efficient Wearable Foundation Model for Women's Health
  5. 2024Wearable Accelerometer Foundation Models for Health via Knowledge Distillation
  6. 2026SOTER: A Generative Time-Series Foundation Model for Wearable Human Physiological Signals
  7. 2025Reducing Annotation Burden in Physical Activity Research Using Vision-Language Models
  8. 2026HEARTS: Benchmarking LLM Reasoning on Health Time Series