@inproceedings{cheng-etal-2024-precision,
title = "Precision or Recall? An Analysis of Image Captions for Training Text-to-Image Generation Model",
author = "Cheng, Sheng and
Patel, Maitreya and
Yang, Yezhou",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-emnlp.211",
pages = "3703--3709",
abstract = "Despite advancements in text-to-image models, generating images that precisely align with textual descriptions remains challenging due to misalignment in training data. In this paper, we analyze the critical role of caption precision and recall in text-to-image model training. Our analysis of human-annotated captions shows that both precision and recall are important for text-image alignment, but precision has a more significant impact. Leveraging these insights, we utilize Large Vision Language Models to generate synthetic captions for training. Models trained with these synthetic captions show similar behavior to those trained on human-annotated captions, underscores the potential for synthetic data in text-to-image training.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cheng-etal-2024-precision">
<titleInfo>
<title>Precision or Recall? An Analysis of Image Captions for Training Text-to-Image Generation Model</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sheng</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maitreya</namePart>
<namePart type="family">Patel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yezhou</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Despite advancements in text-to-image models, generating images that precisely align with textual descriptions remains challenging due to misalignment in training data. In this paper, we analyze the critical role of caption precision and recall in text-to-image model training. Our analysis of human-annotated captions shows that both precision and recall are important for text-image alignment, but precision has a more significant impact. Leveraging these insights, we utilize Large Vision Language Models to generate synthetic captions for training. Models trained with these synthetic captions show similar behavior to those trained on human-annotated captions, underscores the potential for synthetic data in text-to-image training.</abstract>
<identifier type="citekey">cheng-etal-2024-precision</identifier>
<location>
<url>https://aclanthology.org/2024.findings-emnlp.211</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>3703</start>
<end>3709</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Precision or Recall? An Analysis of Image Captions for Training Text-to-Image Generation Model
%A Cheng, Sheng
%A Patel, Maitreya
%A Yang, Yezhou
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F cheng-etal-2024-precision
%X Despite advancements in text-to-image models, generating images that precisely align with textual descriptions remains challenging due to misalignment in training data. In this paper, we analyze the critical role of caption precision and recall in text-to-image model training. Our analysis of human-annotated captions shows that both precision and recall are important for text-image alignment, but precision has a more significant impact. Leveraging these insights, we utilize Large Vision Language Models to generate synthetic captions for training. Models trained with these synthetic captions show similar behavior to those trained on human-annotated captions, underscores the potential for synthetic data in text-to-image training.
%U https://aclanthology.org/2024.findings-emnlp.211
%P 3703-3709
Markdown (Informal)
[Precision or Recall? An Analysis of Image Captions for Training Text-to-Image Generation Model](https://aclanthology.org/2024.findings-emnlp.211) (Cheng et al., Findings 2024)
ACL