2 papers
cs.CV2024
COSMO: COntrastive Streamlined MultimOdal Model with Interleaved Pre-Training
Alex Jinpeng Wang, Linjie Li, Kevin Qinghong Lin +5
In the evolution of Vision-Language Pre-training, shifting from short-text comprehension to encompassing extended textual contexts is pivotal. Recent autoregressive vision-language…
cs.CV2022
Egocentric Video-Language Pretraining @ Ego4D Challenge 2022
Kevin Qinghong Lin, Alex Jinpeng Wang, Mattia Soldan +13
In this report, we propose a video-language pretraining (VLP) based solution \cite{kevin2022egovlp} for four Ego4D challenge tasks, including Natural Language Query (NLQ), Moment Q…