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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Fully fine-tuning pretrained large-scale transformer models has become apopular paradigm for video-language modeling tasks, such as temporal languagegrounding and video-language summarization. With a growing number of tasks andlimited training data, such full fine-tuning approach leads to costly modelstorage and unstable training. To overcome these shortcomings, we introducelightweight adapters to the pre-trained model and only update them atfine-tuning time. However, existing adapters fail to capture intrinsic temporalrelations among video frames or textual words. Moreover, they neglect thepreservation of critical task-related information that flows from the rawvideo-language input into the adapter's low-dimensional space. To address theseissues, we first propose a novel REcurrent ADapter (READ) that employsrecurrent computation to enable temporal modeling capability. Second, wepropose Partial Video-Language Alignment (PVLA) objective via the use ofpartial optimal transport to maintain task-related information flowing into ourREAD modules. We validate our READ-PVLA framework through extensive experimentswhere READ-PVLA significantly outperforms all existing fine-tuning strategieson multiple low-resource temporal language grounding and video-languagesummarization benchmarks.

READ-PVLA: Recurrent Adapter with Partial Video-Language Alignment for Parameter-Efficient Transfer Learning in Low-Resource Video-Language Modeling