On-Device Language Models for Privacy-Preserving Stress Prediction: A Multimodal Evaluation on Mobile Health
TL;DR
This work evaluates ODLMs for multi-modal stress prediction using zero-shot prompting, measuring predictive accuracy alongside latency and throughput, and shows that objective sensor features marginally outperform subjective self-reports on average and that lightweight sub-2B models achieve low latency with predictable resource usage.
Abstract
Stress is a pervasive determinant of mental health and a key target for mobile health interventions. On-device language models (ODLMs) offer privacy-preserving inference without cloud dependency, yet their feasibility for health prediction under mobile resource constraints remains underexplored. We evaluate ODLMs for multi-modal stress prediction using zero-shot prompting, measuring predictive accuracy alongside latency and throughput. Our results show that objective sensor features marginally outperform subjective self-reports on average, and that lightweight sub-2B models achieve low latency with predictable resource usage. Our findings highlight both the promise and the practical constraints of ODLMs for mobile mental health.
