Fetching the paper…

LiteVL: Efficient Video-Language Learning with Enhanced Spatial-Temporal Modeling · Around