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

Source-Free Domain Adaptation (SFDA) aims to adapt a source model for atarget domain, with only access to unlabeled target training data and thesource model pre-trained on a supervised source domain. Relying on pseudolabeling and/or auxiliary supervision, conventional methods are inevitablyerror-prone. To mitigate this limitation, in this work we for the first timeexplore the potentials of off-the-shelf vision-language (ViL) multimodal models(e.g.,CLIP) with rich whilst heterogeneous knowledge. We find that directlyapplying the ViL model to the target domain in a zero-shot fashion isunsatisfactory, as it is not specialized for this particular task but largelygeneric. To make it task specific, we propose a novel Distilling multimodalFoundation model(DIFO)approach. Specifically, DIFO alternates between two stepsduring adaptation: (i) Customizing the ViL model by maximizing the mutualinformation with the target model in a prompt learning manner, (ii) Distillingthe knowledge of this customized ViL model to the target model. For morefine-grained and reliable distillation, we further introduce two effectiveregularization terms, namely most-likely category encouragement and predictiveconsistency. Extensive experiments show that DIFO significantly outperforms thestate-of-the-art alternatives. Our source code will be released.

Source-Free Domain Adaptation with Frozen Multimodal Foundation Model