@inproceedings{zhu-etal-2023-imagine,
title = "{I}magin{E}: An Imagination-Based Automatic Evaluation Metric for Natural Language Generation",
author = "Zhu, Wanrong and
Wang, Xin and
Yan, An and
Eckstein, Miguel and
Wang, William Yang",
editor = "Vlachos, Andreas and
Augenstein, Isabelle",
booktitle = "Findings of the Association for Computational Linguistics: EACL 2023",
month = may,
year = "2023",
address = "Dubrovnik, Croatia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-eacl.6",
doi = "10.18653/v1/2023.findings-eacl.6",
pages = "93--105",
abstract = "Automatic evaluations for natural language generation (NLG) conventionally rely on token-level or embedding-level comparisons with text references. This differs from human language processing, for which visual imagination often improves comprehension. In this work, we propose ImaginE, an imagination-based automatic evaluation metric for natural language generation. With the help of StableDiffusion, a state-of-the-art text-to-image generator, we automatically generate an image as the embodied imagination for the text snippet and compute the imagination similarity using contextual embeddings. Experiments spanning several text generation tasks demonstrate that adding machine-generated images with our ImaginE displays great potential in introducing multi-modal information into NLG evaluation, and improves existing automatic metrics{'} correlations with human similarity judgments in both reference-based and reference-free evaluation scenarios.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhu-etal-2023-imagine">
<titleInfo>
<title>ImaginE: An Imagination-Based Automatic Evaluation Metric for Natural Language Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanrong</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">An</namePart>
<namePart type="family">Yan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Miguel</namePart>
<namePart type="family">Eckstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">William</namePart>
<namePart type="given">Yang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Isabelle</namePart>
<namePart type="family">Augenstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dubrovnik, Croatia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Automatic evaluations for natural language generation (NLG) conventionally rely on token-level or embedding-level comparisons with text references. This differs from human language processing, for which visual imagination often improves comprehension. In this work, we propose ImaginE, an imagination-based automatic evaluation metric for natural language generation. With the help of StableDiffusion, a state-of-the-art text-to-image generator, we automatically generate an image as the embodied imagination for the text snippet and compute the imagination similarity using contextual embeddings. Experiments spanning several text generation tasks demonstrate that adding machine-generated images with our ImaginE displays great potential in introducing multi-modal information into NLG evaluation, and improves existing automatic metrics’ correlations with human similarity judgments in both reference-based and reference-free evaluation scenarios.</abstract>
<identifier type="citekey">zhu-etal-2023-imagine</identifier>
<identifier type="doi">10.18653/v1/2023.findings-eacl.6</identifier>
<location>
<url>https://aclanthology.org/2023.findings-eacl.6</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>93</start>
<end>105</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ImaginE: An Imagination-Based Automatic Evaluation Metric for Natural Language Generation
%A Zhu, Wanrong
%A Wang, Xin
%A Yan, An
%A Eckstein, Miguel
%A Wang, William Yang
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Findings of the Association for Computational Linguistics: EACL 2023
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F zhu-etal-2023-imagine
%X Automatic evaluations for natural language generation (NLG) conventionally rely on token-level or embedding-level comparisons with text references. This differs from human language processing, for which visual imagination often improves comprehension. In this work, we propose ImaginE, an imagination-based automatic evaluation metric for natural language generation. With the help of StableDiffusion, a state-of-the-art text-to-image generator, we automatically generate an image as the embodied imagination for the text snippet and compute the imagination similarity using contextual embeddings. Experiments spanning several text generation tasks demonstrate that adding machine-generated images with our ImaginE displays great potential in introducing multi-modal information into NLG evaluation, and improves existing automatic metrics’ correlations with human similarity judgments in both reference-based and reference-free evaluation scenarios.
%R 10.18653/v1/2023.findings-eacl.6
%U https://aclanthology.org/2023.findings-eacl.6
%U https://doi.org/10.18653/v1/2023.findings-eacl.6
%P 93-105
Markdown (Informal)
[ImaginE: An Imagination-Based Automatic Evaluation Metric for Natural Language Generation](https://aclanthology.org/2023.findings-eacl.6) (Zhu et al., Findings 2023)
ACL