@inproceedings{shao-etal-2020-examination,
title = "Examination and Extension of Strategies for Improving Personalized Language Modeling via Interpolation",
author = "Shao, Liqun and
Mantravadi, Sahitya and
Manzini, Tom and
Buendia, Alejandro and
Knoertzer, Manon and
Srinivasan, Soundar and
Quirk, Chris",
editor = "Awadallah, Ahmed Hassan and
Su, Yu and
Sun, Huan and
Yih, Scott Wen-tau",
booktitle = "Proceedings of the First Workshop on Natural Language Interfaces",
month = jul,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.nli-1.3",
doi = "10.18653/v1/2020.nli-1.3",
pages = "20--26",
abstract = "In this paper, we detail novel strategies for interpolating personalized language models and methods to handle out-of-vocabulary (OOV) tokens to improve personalized language models. Using publicly available data from Reddit, we demonstrate improvements in offline metrics at the user level by interpolating a global LSTM-based authoring model with a user-personalized n-gram model. By optimizing this approach with a back-off to uniform OOV penalty and the interpolation coefficient, we observe that over 80{\%} of users receive a lift in perplexity, with an average of 5.4{\%} in perplexity lift per user. In doing this research we extend previous work in building NLIs and improve the robustness of metrics for downstream tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shao-etal-2020-examination">
<titleInfo>
<title>Examination and Extension of Strategies for Improving Personalized Language Modeling via Interpolation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Liqun</namePart>
<namePart type="family">Shao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sahitya</namePart>
<namePart type="family">Mantravadi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tom</namePart>
<namePart type="family">Manzini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alejandro</namePart>
<namePart type="family">Buendia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manon</namePart>
<namePart type="family">Knoertzer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Soundar</namePart>
<namePart type="family">Srinivasan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chris</namePart>
<namePart type="family">Quirk</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the First Workshop on Natural Language Interfaces</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ahmed</namePart>
<namePart type="given">Hassan</namePart>
<namePart type="family">Awadallah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Huan</namePart>
<namePart type="family">Sun</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">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In this paper, we detail novel strategies for interpolating personalized language models and methods to handle out-of-vocabulary (OOV) tokens to improve personalized language models. Using publicly available data from Reddit, we demonstrate improvements in offline metrics at the user level by interpolating a global LSTM-based authoring model with a user-personalized n-gram model. By optimizing this approach with a back-off to uniform OOV penalty and the interpolation coefficient, we observe that over 80% of users receive a lift in perplexity, with an average of 5.4% in perplexity lift per user. In doing this research we extend previous work in building NLIs and improve the robustness of metrics for downstream tasks.</abstract>
<identifier type="citekey">shao-etal-2020-examination</identifier>
<identifier type="doi">10.18653/v1/2020.nli-1.3</identifier>
<location>
<url>https://aclanthology.org/2020.nli-1.3</url>
</location>
<part>
<date>2020-07</date>
<extent unit="page">
<start>20</start>
<end>26</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Examination and Extension of Strategies for Improving Personalized Language Modeling via Interpolation
%A Shao, Liqun
%A Mantravadi, Sahitya
%A Manzini, Tom
%A Buendia, Alejandro
%A Knoertzer, Manon
%A Srinivasan, Soundar
%A Quirk, Chris
%Y Awadallah, Ahmed Hassan
%Y Su, Yu
%Y Sun, Huan
%Y Yih, Scott Wen-tau
%S Proceedings of the First Workshop on Natural Language Interfaces
%D 2020
%8 July
%I Association for Computational Linguistics
%C Online
%F shao-etal-2020-examination
%X In this paper, we detail novel strategies for interpolating personalized language models and methods to handle out-of-vocabulary (OOV) tokens to improve personalized language models. Using publicly available data from Reddit, we demonstrate improvements in offline metrics at the user level by interpolating a global LSTM-based authoring model with a user-personalized n-gram model. By optimizing this approach with a back-off to uniform OOV penalty and the interpolation coefficient, we observe that over 80% of users receive a lift in perplexity, with an average of 5.4% in perplexity lift per user. In doing this research we extend previous work in building NLIs and improve the robustness of metrics for downstream tasks.
%R 10.18653/v1/2020.nli-1.3
%U https://aclanthology.org/2020.nli-1.3
%U https://doi.org/10.18653/v1/2020.nli-1.3
%P 20-26
Markdown (Informal)
[Examination and Extension of Strategies for Improving Personalized Language Modeling via Interpolation](https://aclanthology.org/2020.nli-1.3) (Shao et al., NLI 2020)
ACL