@inproceedings{matero-etal-2021-melt-message,
title = "{M}e{LT}: Message-Level Transformer with Masked Document Representations as Pre-Training for Stance Detection",
author = "Matero, Matthew and
Soni, Nikita and
Balasubramanian, Niranjan and
Schwartz, H. Andrew",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
month = nov,
year = "2021",
address = "Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.findings-emnlp.253",
doi = "10.18653/v1/2021.findings-emnlp.253",
pages = "2959--2966",
abstract = "Much of natural language processing is focused on leveraging large capacity language models, typically trained over single messages with a task of predicting one or more tokens. However, modeling human language at higher-levels of context (i.e., sequences of messages) is under-explored. In stance detection and other social media tasks where the goal is to predict an attribute of a message, we have contextual data that is loosely semantically connected by authorship. Here, we introduce Message-Level Transformer (MeLT) {--} a hierarchical message-encoder pre-trained over Twitter and applied to the task of stance prediction. We focus on stance prediction as a task benefiting from knowing the context of the message (i.e., the sequence of previous messages). The model is trained using a variant of masked-language modeling; where instead of predicting tokens, it seeks to generate an entire masked (aggregated) message vector via reconstruction loss. We find that applying this pre-trained masked message-level transformer to the downstream task of stance detection achieves F1 performance of 67{\%}.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="matero-etal-2021-melt-message">
<titleInfo>
<title>MeLT: Message-Level Transformer with Masked Document Representations as Pre-Training for Stance Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="family">Matero</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikita</namePart>
<namePart type="family">Soni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Niranjan</namePart>
<namePart type="family">Balasubramanian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">H</namePart>
<namePart type="given">Andrew</namePart>
<namePart type="family">Schwartz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2021</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marie-Francine</namePart>
<namePart type="family">Moens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="given">Wen-tau</namePart>
<namePart type="family">Yih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Punta Cana, Dominican Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Much of natural language processing is focused on leveraging large capacity language models, typically trained over single messages with a task of predicting one or more tokens. However, modeling human language at higher-levels of context (i.e., sequences of messages) is under-explored. In stance detection and other social media tasks where the goal is to predict an attribute of a message, we have contextual data that is loosely semantically connected by authorship. Here, we introduce Message-Level Transformer (MeLT) – a hierarchical message-encoder pre-trained over Twitter and applied to the task of stance prediction. We focus on stance prediction as a task benefiting from knowing the context of the message (i.e., the sequence of previous messages). The model is trained using a variant of masked-language modeling; where instead of predicting tokens, it seeks to generate an entire masked (aggregated) message vector via reconstruction loss. We find that applying this pre-trained masked message-level transformer to the downstream task of stance detection achieves F1 performance of 67%.</abstract>
<identifier type="citekey">matero-etal-2021-melt-message</identifier>
<identifier type="doi">10.18653/v1/2021.findings-emnlp.253</identifier>
<location>
<url>https://aclanthology.org/2021.findings-emnlp.253</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>2959</start>
<end>2966</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T MeLT: Message-Level Transformer with Masked Document Representations as Pre-Training for Stance Detection
%A Matero, Matthew
%A Soni, Nikita
%A Balasubramanian, Niranjan
%A Schwartz, H. Andrew
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Findings of the Association for Computational Linguistics: EMNLP 2021
%D 2021
%8 November
%I Association for Computational Linguistics
%C Punta Cana, Dominican Republic
%F matero-etal-2021-melt-message
%X Much of natural language processing is focused on leveraging large capacity language models, typically trained over single messages with a task of predicting one or more tokens. However, modeling human language at higher-levels of context (i.e., sequences of messages) is under-explored. In stance detection and other social media tasks where the goal is to predict an attribute of a message, we have contextual data that is loosely semantically connected by authorship. Here, we introduce Message-Level Transformer (MeLT) – a hierarchical message-encoder pre-trained over Twitter and applied to the task of stance prediction. We focus on stance prediction as a task benefiting from knowing the context of the message (i.e., the sequence of previous messages). The model is trained using a variant of masked-language modeling; where instead of predicting tokens, it seeks to generate an entire masked (aggregated) message vector via reconstruction loss. We find that applying this pre-trained masked message-level transformer to the downstream task of stance detection achieves F1 performance of 67%.
%R 10.18653/v1/2021.findings-emnlp.253
%U https://aclanthology.org/2021.findings-emnlp.253
%U https://doi.org/10.18653/v1/2021.findings-emnlp.253
%P 2959-2966
Markdown (Informal)
[MeLT: Message-Level Transformer with Masked Document Representations as Pre-Training for Stance Detection](https://aclanthology.org/2021.findings-emnlp.253) (Matero et al., Findings 2021)
ACL