@inproceedings{du-etal-2021-meta,
title = "Meta-Learning with Variational Semantic Memory for Word Sense Disambiguation",
author = "Du, Yingjun and
Holla, Nithin and
Zhen, Xiantong and
Snoek, Cees and
Shutova, Ekaterina",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.409",
doi = "10.18653/v1/2021.acl-long.409",
pages = "5254--5268",
abstract = "A critical challenge faced by supervised word sense disambiguation (WSD) is the lack of large annotated datasets with sufficient coverage of words in their diversity of senses. This inspired recent research on few-shot WSD using meta-learning. While such work has successfully applied meta-learning to learn new word senses from very few examples, its performance still lags behind its fully-supervised counterpart. Aiming to further close this gap, we propose a model of semantic memory for WSD in a meta-learning setting. Semantic memory encapsulates prior experiences seen throughout the lifetime of the model, which aids better generalization in limited data settings. Our model is based on hierarchical variational inference and incorporates an adaptive memory update rule via a hypernetwork. We show our model advances the state of the art in few-shot WSD, supports effective learning in extremely data scarce (e.g. one-shot) scenarios and produces meaning prototypes that capture similar senses of distinct words.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="du-etal-2021-meta">
<titleInfo>
<title>Meta-Learning with Variational Semantic Memory for Word Sense Disambiguation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yingjun</namePart>
<namePart type="family">Du</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nithin</namePart>
<namePart type="family">Holla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiantong</namePart>
<namePart type="family">Zhen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cees</namePart>
<namePart type="family">Snoek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)</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>A critical challenge faced by supervised word sense disambiguation (WSD) is the lack of large annotated datasets with sufficient coverage of words in their diversity of senses. This inspired recent research on few-shot WSD using meta-learning. While such work has successfully applied meta-learning to learn new word senses from very few examples, its performance still lags behind its fully-supervised counterpart. Aiming to further close this gap, we propose a model of semantic memory for WSD in a meta-learning setting. Semantic memory encapsulates prior experiences seen throughout the lifetime of the model, which aids better generalization in limited data settings. Our model is based on hierarchical variational inference and incorporates an adaptive memory update rule via a hypernetwork. We show our model advances the state of the art in few-shot WSD, supports effective learning in extremely data scarce (e.g. one-shot) scenarios and produces meaning prototypes that capture similar senses of distinct words.</abstract>
<identifier type="citekey">du-etal-2021-meta</identifier>
<identifier type="doi">10.18653/v1/2021.acl-long.409</identifier>
<location>
<url>https://aclanthology.org/2021.acl-long.409</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>5254</start>
<end>5268</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Meta-Learning with Variational Semantic Memory for Word Sense Disambiguation
%A Du, Yingjun
%A Holla, Nithin
%A Zhen, Xiantong
%A Snoek, Cees
%A Shutova, Ekaterina
%S Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
%D 2021
%8 August
%I Association for Computational Linguistics
%C Online
%F du-etal-2021-meta
%X A critical challenge faced by supervised word sense disambiguation (WSD) is the lack of large annotated datasets with sufficient coverage of words in their diversity of senses. This inspired recent research on few-shot WSD using meta-learning. While such work has successfully applied meta-learning to learn new word senses from very few examples, its performance still lags behind its fully-supervised counterpart. Aiming to further close this gap, we propose a model of semantic memory for WSD in a meta-learning setting. Semantic memory encapsulates prior experiences seen throughout the lifetime of the model, which aids better generalization in limited data settings. Our model is based on hierarchical variational inference and incorporates an adaptive memory update rule via a hypernetwork. We show our model advances the state of the art in few-shot WSD, supports effective learning in extremely data scarce (e.g. one-shot) scenarios and produces meaning prototypes that capture similar senses of distinct words.
%R 10.18653/v1/2021.acl-long.409
%U https://aclanthology.org/2021.acl-long.409
%U https://doi.org/10.18653/v1/2021.acl-long.409
%P 5254-5268
Markdown (Informal)
[Meta-Learning with Variational Semantic Memory for Word Sense Disambiguation](https://aclanthology.org/2021.acl-long.409) (Du et al., ACL-IJCNLP 2021)
ACL
- Yingjun Du, Nithin Holla, Xiantong Zhen, Cees Snoek, and Ekaterina Shutova. 2021. Meta-Learning with Variational Semantic Memory for Word Sense Disambiguation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 5254–5268, Online. Association for Computational Linguistics.