@inproceedings{zhu-etal-2020-examining,
title = "Examining the rhetorical capacities of neural language models",
author = "Zhu, Zining and
Pan, Chuer and
Abdalla, Mohamed and
Rudzicz, Frank",
booktitle = "Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.blackboxnlp-1.3",
doi = "10.18653/v1/2020.blackboxnlp-1.3",
pages = "16--32",
abstract = "Recently, neural language models (LMs) have demonstrated impressive abilities in generating high-quality discourse. While many recent papers have analyzed the syntactic aspects encoded in LMs, there has been no analysis to date of the inter-sentential, rhetorical knowledge. In this paper, we propose a method that quantitatively evaluates the rhetorical capacities of neural LMs. We examine the capacities of neural LMs understanding the rhetoric of discourse by evaluating their abilities to encode a set of linguistic features derived from Rhetorical Structure Theory (RST). Our experiments show that BERT-based LMs outperform other Transformer LMs, revealing the richer discourse knowledge in their intermediate layer representations. In addition, GPT-2 and XLNet apparently encode less rhetorical knowledge, and we suggest an explanation drawing from linguistic philosophy. Our method shows an avenue towards quantifying the rhetorical capacities of neural LMs.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhu-etal-2020-examining">
<titleInfo>
<title>Examining the rhetorical capacities of neural language models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zining</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chuer</namePart>
<namePart type="family">Pan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="family">Abdalla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Frank</namePart>
<namePart type="family">Rudzicz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recently, neural language models (LMs) have demonstrated impressive abilities in generating high-quality discourse. While many recent papers have analyzed the syntactic aspects encoded in LMs, there has been no analysis to date of the inter-sentential, rhetorical knowledge. In this paper, we propose a method that quantitatively evaluates the rhetorical capacities of neural LMs. We examine the capacities of neural LMs understanding the rhetoric of discourse by evaluating their abilities to encode a set of linguistic features derived from Rhetorical Structure Theory (RST). Our experiments show that BERT-based LMs outperform other Transformer LMs, revealing the richer discourse knowledge in their intermediate layer representations. In addition, GPT-2 and XLNet apparently encode less rhetorical knowledge, and we suggest an explanation drawing from linguistic philosophy. Our method shows an avenue towards quantifying the rhetorical capacities of neural LMs.</abstract>
<identifier type="citekey">zhu-etal-2020-examining</identifier>
<identifier type="doi">10.18653/v1/2020.blackboxnlp-1.3</identifier>
<location>
<url>https://aclanthology.org/2020.blackboxnlp-1.3</url>
</location>
<part>
<date>2020-11</date>
<extent unit="page">
<start>16</start>
<end>32</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Examining the rhetorical capacities of neural language models
%A Zhu, Zining
%A Pan, Chuer
%A Abdalla, Mohamed
%A Rudzicz, Frank
%S Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP
%D 2020
%8 November
%I Association for Computational Linguistics
%C Online
%F zhu-etal-2020-examining
%X Recently, neural language models (LMs) have demonstrated impressive abilities in generating high-quality discourse. While many recent papers have analyzed the syntactic aspects encoded in LMs, there has been no analysis to date of the inter-sentential, rhetorical knowledge. In this paper, we propose a method that quantitatively evaluates the rhetorical capacities of neural LMs. We examine the capacities of neural LMs understanding the rhetoric of discourse by evaluating their abilities to encode a set of linguistic features derived from Rhetorical Structure Theory (RST). Our experiments show that BERT-based LMs outperform other Transformer LMs, revealing the richer discourse knowledge in their intermediate layer representations. In addition, GPT-2 and XLNet apparently encode less rhetorical knowledge, and we suggest an explanation drawing from linguistic philosophy. Our method shows an avenue towards quantifying the rhetorical capacities of neural LMs.
%R 10.18653/v1/2020.blackboxnlp-1.3
%U https://aclanthology.org/2020.blackboxnlp-1.3
%U https://doi.org/10.18653/v1/2020.blackboxnlp-1.3
%P 16-32
Markdown (Informal)
[Examining the rhetorical capacities of neural language models](https://aclanthology.org/2020.blackboxnlp-1.3) (Zhu et al., BlackboxNLP 2020)
ACL