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

Visual recognition in low-data regimes requires deep neural networks to learngeneralized representations from limited training samples. Recently, CLIP-basedmethods have shown promising few-shot performance benefited from thecontrastive language-image pre-training. We then question, if the more diversepre-training knowledge can be cascaded to further assist few-shotrepresentation learning. In this paper, we propose CaFo, a Cascade ofFoundation models that incorporates diverse prior knowledge of variouspre-training paradigms for better few-shot learning. Our CaFo incorporatesCLIP's language-contrastive knowledge, DINO's vision-contrastive knowledge,DALL-E's vision-generative knowledge, and GPT-3's language-generativeknowledge. Specifically, CaFo works by 'Prompt, Generate, then Cache'. Firstly,we leverage GPT-3 to produce textual inputs for prompting CLIP with richdownstream linguistic semantics. Then, we generate synthetic images via DALL-Eto expand the few-shot training data without any manpower. At last, weintroduce a learnable cache model to adaptively blend the predictions from CLIPand DINO. By such collaboration, CaFo can fully unleash the potential ofdifferent pre-training methods and unify them to perform state-of-the-art forfew-shot classification. Code is available athttps://github.com/ZrrSkywalker/CaFo.

Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learners