BALMS is a benchmark for evaluating large language model (LLM) agents that analyze longitudinal wearable data to predict mental‑health wellbeing scores and generate evidence‑grounded rationales. It covers three real‑world datasets, two task families (score prediction and rationale generation), and tests five LLM backbones across open‑ and closed‑source paradigms. The study finds that zero‑shot agents rarely beat a simple mean baseline, and while chain‑of‑thought prompting helps reasoning, it does not ensure temporal grounding or numerical accuracy.
arXiv:2605. 22759v2 Announce Type: replace Abstract: While ubiquitous wearable sensors capture a wealth of behavioral and physiological information, effectively transforming these signals into personalized health insights is challenging.
arXiv:2605. 29483v2 Announce Type: replace Abstract: Wearable devices enable continuous monitoring of physiological signals such as ECG and PPG, but existing mHealth systems are largely limited to task-specific prediction pipelines or reactive question answering over static summaries.
arXiv:2607. 13940v1 Announce Type: new Abstract: Personal health management unfolds over repeated encounters, yet most health AI systems treat each request in isolation.
arXiv:2603. 06638v3 Announce Type: replace-cross Abstract: The rise of large language models (LLMs) has shifted time series analysis from narrow analytics to general-purpose reasoning.
The study introduces a clinician‑in‑the‑loop benchmark to assess whether large language models can generate evidence‑grounded Brief Hierarchical Taxonomy of Psychopathology (B‑HiTOP) item profiles from multimodal data, including passive sensing, ecological momentary assessment, and questionnaires. Using the GLOBEM dataset, the authors create 14,592 participant‑day instances aligned to 29 B‑HiTOP items across five spectra, and evaluate evidence compatibility rather than diagnostic accuracy. Two‑stage prediction improves compatibility for EMA and questionnaire evidence but reduces it for passive sensing and combined evidence, yielding more conservative score distributions across models, spectra, and evidence settings.
arXiv:2601.13880v2 Announce Type: replace Abstract: Personalized lifestyle health analysis requires long-horizon, multi-dimensional reasoning over heterogeneous lifestyle signals, and recent advances...
arXiv:2608. 07518v1 Announce Type: cross Abstract: Longitudinal, in-the-wild, wearable sensing yields day-level physiology, sleep, activity, and environmental streams, whereas affect and cognition are labeled only episodically (per waves).
arXiv:2605. 17679v2 Announce Type: replace-cross Abstract: Cancer survivors face elevated rates of depression, anxiety, and emotional distress, yet self-report may be unavailable at some moments when support is relevant, a challenge we term the diary paradox.
arXiv:2606. 00345v1 Announce Type: new Abstract: Wearable and mobile sensing technologies enable continuous monitoring of human behavior and health in real-world settings.
arXiv:2606. 18147v1 Announce Type: new Abstract: Language models are remarkably capable at medical question answering, in some cases surpassing the accuracy of general physicians.
arXiv:2606. 14604v1 Announce Type: cross Abstract: Wearable devices and smartphones generate rich behavioural time series that can support proactive health interventions, yet systematic comparisons of modern forecasting architectures for these data are lacking.
arXiv:2607. 09322v1 Announce Type: new Abstract: In this work, we introduce LongMedBench, a real-world EHR-based benchmark for long-horizon clinical decision-making.