@inproceedings{liu-etal-2023-rapid,
title = "Rapid Diffusion: Building Domain-Specific Text-to-Image Synthesizers with Fast Inference Speed",
author = "Liu, Bingyan and
Lin, Weifeng and
Duan, Zhongjie and
Wang, Chengyu and
Ziheng, Wu and
Zipeng, Zhang and
Jia, Kui and
Jin, Lianwen and
Chen, Cen and
Huang, Jun",
editor = "Sitaram, Sunayana and
Beigman Klebanov, Beata and
Williams, Jason D",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-industry.28",
doi = "10.18653/v1/2023.acl-industry.28",
pages = "295--304",
abstract = "Text-to-Image Synthesis (TIS) aims to generate images based on textual inputs. Recently, several large pre-trained diffusion models have been released to create high-quality images with pre-trained text encoders and diffusion-based image synthesizers. However, popular diffusion-based models from the open-source community cannot support industrial domain-specific applications due to the lack of entity knowledge and low inference speed. In this paper, we propose Rapid Diffusion, a novel framework for training and deploying super-resolution, text-to-image latent diffusion models with rich entity knowledge injected and optimized networks. Furthermore, we employ BladeDISC, an end-to-end Artificial Intelligence (AI) compiler, and FlashAttention techniques to optimize computational graphs of the generated models for online deployment. Experiments verify the effectiveness of our approach in terms of image quality and inference speed. In addition, we present industrial use cases and integrate Rapid Diffusion to an AI platform to show its practical values.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2023-rapid">
<titleInfo>
<title>Rapid Diffusion: Building Domain-Specific Text-to-Image Synthesizers with Fast Inference Speed</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bingyan</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weifeng</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhongjie</namePart>
<namePart type="family">Duan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengyu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wu</namePart>
<namePart type="family">Ziheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhang</namePart>
<namePart type="family">Zipeng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kui</namePart>
<namePart type="family">Jia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lianwen</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cen</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sunayana</namePart>
<namePart type="family">Sitaram</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Beata</namePart>
<namePart type="family">Beigman Klebanov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jason</namePart>
<namePart type="given">D</namePart>
<namePart type="family">Williams</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Text-to-Image Synthesis (TIS) aims to generate images based on textual inputs. Recently, several large pre-trained diffusion models have been released to create high-quality images with pre-trained text encoders and diffusion-based image synthesizers. However, popular diffusion-based models from the open-source community cannot support industrial domain-specific applications due to the lack of entity knowledge and low inference speed. In this paper, we propose Rapid Diffusion, a novel framework for training and deploying super-resolution, text-to-image latent diffusion models with rich entity knowledge injected and optimized networks. Furthermore, we employ BladeDISC, an end-to-end Artificial Intelligence (AI) compiler, and FlashAttention techniques to optimize computational graphs of the generated models for online deployment. Experiments verify the effectiveness of our approach in terms of image quality and inference speed. In addition, we present industrial use cases and integrate Rapid Diffusion to an AI platform to show its practical values.</abstract>
<identifier type="citekey">liu-etal-2023-rapid</identifier>
<identifier type="doi">10.18653/v1/2023.acl-industry.28</identifier>
<location>
<url>https://aclanthology.org/2023.acl-industry.28</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>295</start>
<end>304</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Rapid Diffusion: Building Domain-Specific Text-to-Image Synthesizers with Fast Inference Speed
%A Liu, Bingyan
%A Lin, Weifeng
%A Duan, Zhongjie
%A Wang, Chengyu
%A Ziheng, Wu
%A Zipeng, Zhang
%A Jia, Kui
%A Jin, Lianwen
%A Chen, Cen
%A Huang, Jun
%Y Sitaram, Sunayana
%Y Beigman Klebanov, Beata
%Y Williams, Jason D.
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F liu-etal-2023-rapid
%X Text-to-Image Synthesis (TIS) aims to generate images based on textual inputs. Recently, several large pre-trained diffusion models have been released to create high-quality images with pre-trained text encoders and diffusion-based image synthesizers. However, popular diffusion-based models from the open-source community cannot support industrial domain-specific applications due to the lack of entity knowledge and low inference speed. In this paper, we propose Rapid Diffusion, a novel framework for training and deploying super-resolution, text-to-image latent diffusion models with rich entity knowledge injected and optimized networks. Furthermore, we employ BladeDISC, an end-to-end Artificial Intelligence (AI) compiler, and FlashAttention techniques to optimize computational graphs of the generated models for online deployment. Experiments verify the effectiveness of our approach in terms of image quality and inference speed. In addition, we present industrial use cases and integrate Rapid Diffusion to an AI platform to show its practical values.
%R 10.18653/v1/2023.acl-industry.28
%U https://aclanthology.org/2023.acl-industry.28
%U https://doi.org/10.18653/v1/2023.acl-industry.28
%P 295-304
Markdown (Informal)
[Rapid Diffusion: Building Domain-Specific Text-to-Image Synthesizers with Fast Inference Speed](https://aclanthology.org/2023.acl-industry.28) (Liu et al., ACL 2023)
ACL
- Bingyan Liu, Weifeng Lin, Zhongjie Duan, Chengyu Wang, Wu Ziheng, Zhang Zipeng, Kui Jia, Lianwen Jin, Cen Chen, and Jun Huang. 2023. Rapid Diffusion: Building Domain-Specific Text-to-Image Synthesizers with Fast Inference Speed. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track), pages 295–304, Toronto, Canada. Association for Computational Linguistics.