{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/ernie-vilg-unified-generative-pre-training","title":"ERNIE-ViLG: Unified Generative Pre-training for Bidirectional Vision-Language Generation","arxiv_id":"2112.15283","date":"2021-12-31","proceeding":null,"authors":["Han Zhang","Weichong Yin","Yewei Fang","Lanxin Li","Boqiang Duan","Zhihua Wu","Yu Sun","Hao Tian","Hua Wu","Haifeng Wang"],"abstract":"Conventional methods for the image-text generation tasks mainly tackle the naturally bidirectional generation tasks separately, focusing on designing task-specific frameworks to improve the quality and fidelity of the generated samples. Recently, Vision-Language Pre-training models have greatly improved the performance of the image-to-text generation tasks, but large-scale pre-training models for text-to-image synthesis task are still under-developed. In this paper, we propose ERNIE-ViLG, a unified generative pre-training framework for bidirectional image-text generation with transformer model. Based on the image quantization models, we formulate both image generation and text generation as autoregressive generative tasks conditioned on the text/image input. The bidirectional image-text generative modeling eases the semantic alignments across vision and language. For the text-to-image generation process, we further propose an end-to-end training method to jointly learn the visual sequence generator and the image reconstructor. To explore the landscape of large-scale pre-training for bidirectional text-image generation, we train a 10-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 both text-to-image and image-to-text tasks, obtaining an FID of 7.9 on MS-COCO for text-to-image synthesis and best results on COCO-CN and AIC-ICC for image captioning.","url_abs":"https://arxiv.org/abs/2112.15283v1","url_pdf":"https://arxiv.org/pdf/2112.15283v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"ernie-vilg-unified-generative-pre-training","repo_url":"https://github.com/PaddlePaddle/FleetX","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"ernie-vilg-unified-generative-pre-training","repo_url":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/pipelines/examples/text_to_image","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null}],"tasks":[{"task_slug":"image-captioning","task_name":"Image Captioning"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"image-to-text","task_name":"Image to text"},{"task_slug":"quantization","task_name":"Quantization"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"text-to-image-generation-1","task_name":"Text to Image Generation"},{"task_slug":"text-to-image-generation","task_name":"Text-to-Image Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/text-to-image-generation-on-coco","task":"Text-to-Image Generation","dataset":"COCO (Common Objects in Context)","model":"ERNIE-ViLG","rank_in_archive_order":40,"of":69,"metrics":{"FID":"14.7"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2112.15283","atlas_url":"https://app.syntology.ai/?focus=2112.15283","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}