@inproceedings{song-etal-2023-model,
title = "Model Intrinsic Features of Fine-tuning based Text Summarization Models for Factual Consistency",
author = "Song, Jongyoon and
Park, Nohil and
Hwang, Bongkyu and
Yun, Jaewoong and
Joe, Seongho and
Gwon, Youngjune and
Yoon, Sungroh",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.872",
doi = "10.18653/v1/2023.findings-acl.872",
pages = "13884--13898",
abstract = "In this study, we analyze the model intrinsic features of a summarization model by varying the fine-tuning objectives and datasets. We fine-tune BART models combining three fine-tuning objectives (negative log-likelihood, unlikelihood, and contrastive loss) and two datasets (CNN/DailyMail and XSum) and provide shuffled or aligned documents to observe changes in the model predictions and intrinsic features. We find that (i) the inductive bias for factual consistency during the fine-tuning procedure depends on both the objectives and datasets, and (ii) summarization models with relatively low factual consistency are more likely to model summaries that are not conditional to the documents. We demonstrate that splitting data based on the unconditional and conditional summary modeling difficulty affects the factual consistency and intrinsic features of the summarization models. Our experimental results highlight the importance of studying the inductive bias during fine-tuning for factual consistency.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="song-etal-2023-model">
<titleInfo>
<title>Model Intrinsic Features of Fine-tuning based Text Summarization Models for Factual Consistency</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jongyoon</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nohil</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bongkyu</namePart>
<namePart type="family">Hwang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jaewoong</namePart>
<namePart type="family">Yun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seongho</namePart>
<namePart type="family">Joe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Youngjune</namePart>
<namePart type="family">Gwon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sungroh</namePart>
<namePart type="family">Yoon</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>Findings of the Association for Computational Linguistics: ACL 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</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>In this study, we analyze the model intrinsic features of a summarization model by varying the fine-tuning objectives and datasets. We fine-tune BART models combining three fine-tuning objectives (negative log-likelihood, unlikelihood, and contrastive loss) and two datasets (CNN/DailyMail and XSum) and provide shuffled or aligned documents to observe changes in the model predictions and intrinsic features. We find that (i) the inductive bias for factual consistency during the fine-tuning procedure depends on both the objectives and datasets, and (ii) summarization models with relatively low factual consistency are more likely to model summaries that are not conditional to the documents. We demonstrate that splitting data based on the unconditional and conditional summary modeling difficulty affects the factual consistency and intrinsic features of the summarization models. Our experimental results highlight the importance of studying the inductive bias during fine-tuning for factual consistency.</abstract>
<identifier type="citekey">song-etal-2023-model</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.872</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.872</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>13884</start>
<end>13898</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Model Intrinsic Features of Fine-tuning based Text Summarization Models for Factual Consistency
%A Song, Jongyoon
%A Park, Nohil
%A Hwang, Bongkyu
%A Yun, Jaewoong
%A Joe, Seongho
%A Gwon, Youngjune
%A Yoon, Sungroh
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F song-etal-2023-model
%X In this study, we analyze the model intrinsic features of a summarization model by varying the fine-tuning objectives and datasets. We fine-tune BART models combining three fine-tuning objectives (negative log-likelihood, unlikelihood, and contrastive loss) and two datasets (CNN/DailyMail and XSum) and provide shuffled or aligned documents to observe changes in the model predictions and intrinsic features. We find that (i) the inductive bias for factual consistency during the fine-tuning procedure depends on both the objectives and datasets, and (ii) summarization models with relatively low factual consistency are more likely to model summaries that are not conditional to the documents. We demonstrate that splitting data based on the unconditional and conditional summary modeling difficulty affects the factual consistency and intrinsic features of the summarization models. Our experimental results highlight the importance of studying the inductive bias during fine-tuning for factual consistency.
%R 10.18653/v1/2023.findings-acl.872
%U https://aclanthology.org/2023.findings-acl.872
%U https://doi.org/10.18653/v1/2023.findings-acl.872
%P 13884-13898
Markdown (Informal)
[Model Intrinsic Features of Fine-tuning based Text Summarization Models for Factual Consistency](https://aclanthology.org/2023.findings-acl.872) (Song et al., Findings 2023)
ACL