@inproceedings{prange-etal-2022-linguistic,
title = "Linguistic Frameworks Go Toe-to-Toe at Neuro-Symbolic Language Modeling",
author = "Prange, Jakob and
Schneider, Nathan and
Kong, Lingpeng",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.naacl-main.325/",
doi = "10.18653/v1/2022.naacl-main.325",
pages = "4375--4391",
abstract = "We examine the extent to which, in principle, different syntactic and semantic graph representations can complement and improve neural language modeling. Specifically, by conditioning on a subgraph encapsulating the locally relevant sentence history, can a model make better next-word predictions than a pretrained sequential language model alone? With an ensemble setup consisting of GPT-2 and ground-truth graphs from one of 7 different formalisms, we find that the graph information indeed improves perplexity and other metrics. Moreover, this architecture provides a new way to compare different frameworks of linguistic representation. In our oracle graph setup, training and evaluating on English WSJ, semantic constituency structures prove most useful to language modeling performance{---}outpacing syntactic constituency structures as well as syntactic and semantic dependency structures."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="prange-etal-2022-linguistic">
<titleInfo>
<title>Linguistic Frameworks Go Toe-to-Toe at Neuro-Symbolic Language Modeling</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jakob</namePart>
<namePart type="family">Prange</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nathan</namePart>
<namePart type="family">Schneider</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lingpeng</namePart>
<namePart type="family">Kong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marine</namePart>
<namePart type="family">Carpuat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie-Catherine</namePart>
<namePart type="family">de Marneffe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="given">Vladimir</namePart>
<namePart type="family">Meza Ruiz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Seattle, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We examine the extent to which, in principle, different syntactic and semantic graph representations can complement and improve neural language modeling. Specifically, by conditioning on a subgraph encapsulating the locally relevant sentence history, can a model make better next-word predictions than a pretrained sequential language model alone? With an ensemble setup consisting of GPT-2 and ground-truth graphs from one of 7 different formalisms, we find that the graph information indeed improves perplexity and other metrics. Moreover, this architecture provides a new way to compare different frameworks of linguistic representation. In our oracle graph setup, training and evaluating on English WSJ, semantic constituency structures prove most useful to language modeling performance—outpacing syntactic constituency structures as well as syntactic and semantic dependency structures.</abstract>
<identifier type="citekey">prange-etal-2022-linguistic</identifier>
<identifier type="doi">10.18653/v1/2022.naacl-main.325</identifier>
<location>
<url>https://aclanthology.org/2022.naacl-main.325/</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>4375</start>
<end>4391</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Linguistic Frameworks Go Toe-to-Toe at Neuro-Symbolic Language Modeling
%A Prange, Jakob
%A Schneider, Nathan
%A Kong, Lingpeng
%Y Carpuat, Marine
%Y de Marneffe, Marie-Catherine
%Y Meza Ruiz, Ivan Vladimir
%S Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, United States
%F prange-etal-2022-linguistic
%X We examine the extent to which, in principle, different syntactic and semantic graph representations can complement and improve neural language modeling. Specifically, by conditioning on a subgraph encapsulating the locally relevant sentence history, can a model make better next-word predictions than a pretrained sequential language model alone? With an ensemble setup consisting of GPT-2 and ground-truth graphs from one of 7 different formalisms, we find that the graph information indeed improves perplexity and other metrics. Moreover, this architecture provides a new way to compare different frameworks of linguistic representation. In our oracle graph setup, training and evaluating on English WSJ, semantic constituency structures prove most useful to language modeling performance—outpacing syntactic constituency structures as well as syntactic and semantic dependency structures.
%R 10.18653/v1/2022.naacl-main.325
%U https://aclanthology.org/2022.naacl-main.325/
%U https://doi.org/10.18653/v1/2022.naacl-main.325
%P 4375-4391
Markdown (Informal)
[Linguistic Frameworks Go Toe-to-Toe at Neuro-Symbolic Language Modeling](https://aclanthology.org/2022.naacl-main.325/) (Prange et al., NAACL 2022)
ACL
- Jakob Prange, Nathan Schneider, and Lingpeng Kong. 2022. Linguistic Frameworks Go Toe-to-Toe at Neuro-Symbolic Language Modeling. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4375–4391, Seattle, United States. Association for Computational Linguistics.