3 papers indexed
Yunxiao Shi, Hong Cai, Mohammad Ghavamzadeh, Fatih Porikli
End-to-end autonomous driving (E2E-AD) aims to directly map raw sensor information to driving actions. Recently, with the rapid advancement of multi-modal large language models (MLLMs), researchers have proposed the paradigm of Vision-Language-Action (VLA) models for E2E-AD, wher…
Jamie Menjay Lin, Jisoo Jeong, Hong Cai, Kai Wang, Fatih Porikli
Motions of objects and scenes carry essential intelligence in video understanding, offering rich cues for interpreting dynamic settings and interactions. Due to the cost and scarcity of high-quality annotation or ground truth of pixel-wise optical flow, however, motion estimation…
Hong Cai, Dandan Shen, Xiangjun Hu, Hongwei Yin, Zhangren Yan
Ochratoxin A (OTA), a prevalent food contaminant, has been proposed as a potential contributor to the development of prostate cancer, although its precise mechanisms remain unclear. This study employed a comprehensive approach that integrated network toxicology, machine learning,…