Local Ai
On-Device Language Models for Privacy-Preserving Stress Prediction: A Multimodal Evaluation on Mobile Health
arXiv:2609.11961v1 Announce Type: new Abstract: Stress is a pervasive determinant of mental health and a key target for mobile health interventions. On-device language models (ODLMs) offer privacy-pre
arXiv:2609.11961v1 Announce Type: new 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.
Related
- Quant.npu: Enabling Efficient Mobile NPU Inference for on-device LLMs via Fully Static Quantization
- Memory-Efficient Structured Backpropagation for On-Device LLM Fine-Tuning
- LeanStream: A Speculate-and-Refine Streaming Framework for Efficient on-Device LLM Inference
- EnerInfer: Energy-Aware On-Device LLM Inference
Source: arXiv cs.LG | 2026-09-14