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

Conventional methods for the image-text generation tasks mainly tackle thenaturally bidirectional generation tasks separately, focusing on designingtask-specific frameworks to improve the quality and fidelity of the generatedsamples. Recently, Vision-Language Pre-training models have greatly improvedthe performance of the image-to-text generation tasks, but large-scalepre-training models for text-to-image synthesis task are still under-developed.In this paper, we propose ERNIE-ViLG, a unified generative pre-trainingframework for bidirectional image-text generation with transformer model. Basedon the image quantization models, we formulate both image generation and textgeneration as autoregressive generative tasks conditioned on the text/imageinput. The bidirectional image-text generative modeling eases the semanticalignments across vision and language. For the text-to-image generationprocess, we further propose an end-to-end training method to jointly learn thevisual sequence generator and the image reconstructor. To explore the landscapeof large-scale pre-training for bidirectional text-image generation, we train a10-billion parameter ERNIE-ViLG model on a large-scale dataset of 145 million(Chinese) image-text pairs which achieves state-of-the-art performance for bothtext-to-image and image-to-text tasks, obtaining an FID of 7.9 on MS-COCO fortext-to-image synthesis and best results on COCO-CN and AIC-ICC for imagecaptioning.

ERNIE-ViLG: Unified Generative Pre-training for Bidirectional Vision-Language Generation