STELLA: Efficient Sensor-to-LLM Translation for On-Device Human Activity Recognition
About
HAR is increasingly expected to run continuously on edge devices, yet recent LLM-based methods remain hard to deploy: raw sensor prompts are long, cloud inference adds latency and privacy risk, and fine-tuned LLM pipelines turn general-purpose models into task-specific classifiers. We present STELLA, an efficient sensor-to-LLM translation framework for on-device HAR that shifts the burden from LLM adaptation to sensor tokenization. A lightweight hierarchical tokenizer compresses an entire multi-channel inertial window into a fixed set of compact latent sensor tokens, which are projected into the embedding space of a frozen pretrained LLM and combined with a natural-language prompt for label scoring. This preserves activity-relevant temporal and cross-channel structure while keeping LLM-side computation predictable across sensor configurations. STELLA also supports on-device personalization, adapting only the lightweight tokenizer on small amounts of user-specific labelled data and augmenting inference with a local retrieval context, keeping the LLM, user data, and retrieval on device. Across seven public HAR datasets and eight benchmark settings, STELLA achieves new state-of-the-art performance, improving over prior methods by up to 11.83% F1; on-device personalization yields up to a further 21.91% F1 as user data accumulates after deployment. STELLA also outperforms representative time-series tokenizers under the same LLM pipeline and achieves real-time inference under practical mobile and edge budgets, showing that efficient sensor tokenization is a practical path toward accurate, private, and personalized LLM-based HAR on edge devices.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Human Activity Recognition | PAMAP2 | Accuracy96.48 | 66 | |
| Human Activity Recognition | Opportunity | Macro F194.29 | 51 | |
| Activity Recognition | mHealth | F1 Score97.15 | 46 | |
| Human Activity Recognition | HHAR | Accuracy72.71 | 39 | |
| Human Activity Recognition | USC-HAD | Macro F172 | 33 | |
| Human Activity Recognition | GOTOV | F1 Score82.6 | 15 | |
| Human Activity Recognition | SKODA | F1 Score95.87 | 15 | |
| Human Activity Recognition | Opportunity gesture | Accuracy79.88 | 7 |