@inproceedings{vijjini-etal-2023-curricular,
title = "Curricular Next Conversation Prediction Pretraining for Transcript Segmentation",
author = "Vijjini, Anvesh Rao and
Deilamsalehy, Hanieh and
Dernoncourt, Franck and
Chaturvedi, Snigdha",
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.197",
doi = "10.18653/v1/2023.findings-eacl.197",
pages = "2597--2607",
abstract = "Transcript segmentation is the task of dividing a single continuous transcript into multiple segments. While document segmentation is a popular task, transcript segmentation has significant challenges due to the relatively noisy and sporadic nature of data. We propose pretraining strategies to address these challenges. The strategies are based on {``}Next Conversation Prediction{''} (NCP) with the underlying idea of pretraining a model to identify consecutive conversations. We further introduce {``}Advanced NCP{''} to make the pretraining task more relevant to the downstream task of segmentation break prediction while being significantly easier. Finally we introduce a curriculum to Advanced NCP (Curricular NCP) based on the similarity between pretraining and downstream task samples. Curricular NCP applied to a state-of-the-art model for text segmentation outperforms prior results. We also show that our pretraining strategies make the model robust to speech recognition errors commonly found in automatically generated transcripts.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vijjini-etal-2023-curricular">
<titleInfo>
<title>Curricular Next Conversation Prediction Pretraining for Transcript Segmentation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anvesh</namePart>
<namePart type="given">Rao</namePart>
<namePart type="family">Vijjini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanieh</namePart>
<namePart type="family">Deilamsalehy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Franck</namePart>
<namePart type="family">Dernoncourt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Snigdha</namePart>
<namePart type="family">Chaturvedi</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>Transcript segmentation is the task of dividing a single continuous transcript into multiple segments. While document segmentation is a popular task, transcript segmentation has significant challenges due to the relatively noisy and sporadic nature of data. We propose pretraining strategies to address these challenges. The strategies are based on “Next Conversation Prediction” (NCP) with the underlying idea of pretraining a model to identify consecutive conversations. We further introduce “Advanced NCP” to make the pretraining task more relevant to the downstream task of segmentation break prediction while being significantly easier. Finally we introduce a curriculum to Advanced NCP (Curricular NCP) based on the similarity between pretraining and downstream task samples. Curricular NCP applied to a state-of-the-art model for text segmentation outperforms prior results. We also show that our pretraining strategies make the model robust to speech recognition errors commonly found in automatically generated transcripts.</abstract>
<identifier type="citekey">vijjini-etal-2023-curricular</identifier>
<identifier type="doi">10.18653/v1/2023.findings-eacl.197</identifier>
<location>
<url>https://aclanthology.org/2023.findings-eacl.197</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>2597</start>
<end>2607</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Curricular Next Conversation Prediction Pretraining for Transcript Segmentation
%A Vijjini, Anvesh Rao
%A Deilamsalehy, Hanieh
%A Dernoncourt, Franck
%A Chaturvedi, Snigdha
%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 vijjini-etal-2023-curricular
%X Transcript segmentation is the task of dividing a single continuous transcript into multiple segments. While document segmentation is a popular task, transcript segmentation has significant challenges due to the relatively noisy and sporadic nature of data. We propose pretraining strategies to address these challenges. The strategies are based on “Next Conversation Prediction” (NCP) with the underlying idea of pretraining a model to identify consecutive conversations. We further introduce “Advanced NCP” to make the pretraining task more relevant to the downstream task of segmentation break prediction while being significantly easier. Finally we introduce a curriculum to Advanced NCP (Curricular NCP) based on the similarity between pretraining and downstream task samples. Curricular NCP applied to a state-of-the-art model for text segmentation outperforms prior results. We also show that our pretraining strategies make the model robust to speech recognition errors commonly found in automatically generated transcripts.
%R 10.18653/v1/2023.findings-eacl.197
%U https://aclanthology.org/2023.findings-eacl.197
%U https://doi.org/10.18653/v1/2023.findings-eacl.197
%P 2597-2607
Markdown (Informal)
[Curricular Next Conversation Prediction Pretraining for Transcript Segmentation](https://aclanthology.org/2023.findings-eacl.197) (Vijjini et al., Findings 2023)
ACL