DailyLLM: Context-Aware Activity Log Generation Using Multi-Modal Sensors and LLMs
Ye Tian, Xiaoyuan Ren, Zihao Wang, Onat Gungor, Xiaofan Yu, Tajana Rosing
22nd IEEE International Conference on Mobile Ad-Hoc and Smart Systems (MASS), 2025, pp. 372–380.
Published Paper (DOI) · Project Website · Paper PDF
Recommended citation: Tian, Y., Ren, X., Wang, Z., Gungor, O., Yu, X., & Rosing, T. (2025). DailyLLM: Context-Aware Activity Log Generation Using Multi-Modal Sensors and LLMs. IEEE International Conference on Mobile Ad-Hoc and Smart Systems (MASS), pp. 372–380. https://doi.org/10.1109/MASS66014.2025.00060 https://arxiv.org/abs/2507.13737
DailyLLM is a lightweight LLM system for context-aware activity log generation and summarization using multimodal smartphone and smartwatch sensors. It uses multi-task LoRA fine-tuning to balance model capability and system efficiency. The paper reports four minutes to summarize a two-hour activity window on Raspberry Pi 5.