Improving Zero-shot ADL Recognition with Large Language Models through Event-based Context

Unobtrusive sensor-based recognition of Activities of Daily Living (ADLs) in smart homes by processing data collected from IoT sensing devices supports applications such as healthcare, safety, and energy management. Recent zero-shot methods based on Large Language Models (LLMs) remove the reliance on labeled ADL sensor data. However, existing approaches still rely on time-based segmentation, which is poorly aligned with the contextual reasoning capabilities of LLMs. This paper proposes to improve zero-shot ADL recognition with event-based segmentation. Our experimental evaluation shows that event-based segmentation consistently outperforms time-based LLM approaches on complex, realistic datasets and remains competitive with a supervised event-based baseline, even with a relatively small LLM such as Gemma 3 27B.