SOURCE-LINKED INTELLIGENCE

BALMS: Benchmarking Agentic LLMs for Longitudinal Mental Health Sensing

arXiv · AI, language, vision and robotics · article · Aug 27, 2026 · UTC

Mental health assessment relies on episodic self-report scales, which convert subjective states such as stress into numerical scores but provide only sparse snapshots of wellbeing. Wearable devices offer longitudinal behavioral and physiological signals for continuous, low-burden monitoring. Recent LLM-driven personal-health agents enable natural language queries over wearable signals, but mainly handle short-term, retrieval-based lookups (e.g., highest step count over a week). They do not evaluate whether agents can reason over long-term signals to predict wellbeing scores paired with evidenc

Read original source ↗ Open in workspace

recordType

paper

region

Global