English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent large-scale video-language pre-trained models have shown appealingperformance on various downstream tasks. However, the pre-training process iscomputationally expensive due to the requirement of millions of video-textpairs and the redundant data structure of each video. To mitigate theseproblems, we propose LiteVL, which adapts a pre-trained image-language modelBLIP into a video-text model directly on downstream tasks, without heavypre-training. To enhance the temporal modeling lacking in the image-languagemodel, we propose to add temporal attention modules in the image encoder ofBLIP with dynamic temporal scaling. Besides the model-wise adaptation, we alsopropose a non-parametric pooling mechanism to adaptively reweight thefine-grained video embedding conditioned on the text. Experimental results ontext-video retrieval and video question answering show that the proposed LiteVLeven outperforms previous video-language pre-trained models by a clear margin,though without any video-language pre-training.

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