COMODO: Cross-Modal Video-to-IMU Distillation for Efficient Egocentric Human Activity Recognition Paper • 2503.07259 • Published Mar 10
SensorLLM: Human-Intuitive Alignment of Multivariate Sensor Data with LLMs for Activity Recognition Paper • 2410.10624 • Published Oct 14, 2024 • 1
ZARA: Zero-shot Motion Time-Series Analysis via Knowledge and Retrieval Driven LLM Agents Paper • 2508.04038 • Published 8 days ago • 1
Evaluating Dialect Robustness of Language Models via Conversation Understanding Paper • 2405.05688 • Published May 9, 2024