TRACE-TS: Attribution-Grounded and Traceable Sensor-Language Reasoning for Human Activity Understanding

본문 미리보기

arXiv:2608.00200v1 Announce Type: new Abstract: Wearable sensors capture fine-grained motion patterns that support rich behavioral understanding, yet most existing methods reduce these signals to activity labels. Recent LM-based approaches generate natural-language explanations for sensor data, but their reasoning is weakly grounded in the underlying signal, leading to fluent yet unverifiable explanations. We introduce TRACE-TS (Traceable Reasoning with Attribution-Grounded Evidence), a framewo