# Towards a General Intelligence and Interface for Wearable Health Data

arXiv: https://arxiv.org/abs/2605.22759 · HTML: https://mlsanity.com/paper/2605.22759/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 72 references that are in the map.
- First posted on arXiv 21 May 2026.

## Builds on (most cited first, 18 of 18)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [A decoder-only foundation model for time-series forecasting](https://mlsanity.com/paper/2310.10688/) (2023)
- [MOMENT: A Family of Open Time-series Foundation Models](https://mlsanity.com/paper/2402.03885/) (2024)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [TimeGPT-1](https://mlsanity.com/paper/2310.03589/) (2023)
- [Large Language Models are Few-Shot Health Learners](https://mlsanity.com/paper/2305.15525/) (2023)
- [Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting](https://mlsanity.com/paper/2310.08278/) (2023)
- [Language Models Still Struggle to Zero-shot Reason about Time Series](https://mlsanity.com/paper/2404.11757/) (2024)
- [An AI system to help scientists write expert-level empirical software](https://mlsanity.com/paper/2509.06503/) (2025)
- [SleepFM: Multi-modal Representation Learning for Sleep Across Brain Activity, ECG and Respiratory Signals](https://mlsanity.com/paper/2405.17766/) (2024)
- [LSM-2: Learning from Incomplete Wearable Sensor Data](https://mlsanity.com/paper/2506.05321/) (2025)
- [The Anatomy of a Personal Health Agent](https://mlsanity.com/paper/2508.20148/) (2025)
- [SymptomAI: Toward a Conversational AI Agent for Everyday Symptom Assessment](https://mlsanity.com/paper/2605.04012/) (2026)

## Built on by (most cited first, 6 of 6)

- [WEQA: Wearable hEalth Question Answering with Query-Adaptive Agentic Reasoning](https://mlsanity.com/paper/2606.18147/) (2026)
- [OpenMHC: Accelerating the Science of Wearable Foundation Models](https://mlsanity.com/paper/2607.16235/) (2026)
- [Kepler-Encoder-v0.1: Towards a Multimodal Embedding Model for Robots](https://mlsanity.com/paper/2607.13522/) (2026)
- [Capturing Cardiac Cyclicity through Phase-Equivariant Self-Supervised Learning](https://mlsanity.com/paper/2608.21147/) (2026)
- [Learning Human Health and Diseases from 24-hour Wrist Movement](https://mlsanity.com/paper/2608.29494/) (2026)
- [WearableQA: A Benchmark for Health Reasoning over Real-World Wearable Data](https://mlsanity.com/paper/2609.05405/) (2026)

## Abstract (arXiv)

While ubiquitous wearable sensors capture a wealth of behavioral and physiological information, effectively transforming these signals into personalized health insights is challenging. Specifically, converting low-level sensor data into representations capable of characterizing higher-level states is difficult due to high phenotypic diversity and variation in individual baseline health, physiology, and lifestyle factors. Moreover, collecting wearable data paired with health outcome annotations is laborious and expensive, and retrospective annotation remains practically unfeasible, contributing to a scarcity of data with high-quality labels. To overcome these limitations, we propose a foundation model for wearable health that is pretrained on more than one trillion minutes of unlabeled sensor signals drawn from a large cohort of five million participants. We demonstrate that the joint scaling of model capacity and pretraining data volume leads to systematic improvements in performance, as evaluated on a diverse set of 35 health prediction tasks, spanning cardiovascular, metabolic, sleep, and mental health, as well as lifestyle choices and demographic factors. We find that this population scale representation unlocks label-efficient few-shot learning and generative capabilities for robust daily metric estimation. To further leverage this learned representation, we deploy a classroom of LLM agents to autonomously search the space of downstream predictive heads built on the model embeddings, showing broad performance improvements that increase with LLM model capacity. Finally, we show how integrating these downstream predictors into a Personal Health Agent can support model responses that are more relevant, contextually aware, and safe, and we validate this via 1,860 ratings from a cohort of clinicians.