@inproceedings{xie-etal-2022-wr,
title = "{WR}-{O}ne2{S}et: Towards Well-Calibrated Keyphrase Generation",
author = "Xie, Binbin and
Wei, Xiangpeng and
Yang, Baosong and
Lin, Huan and
Xie, Jun and
Wang, Xiaoli and
Zhang, Min and
Su, Jinsong",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.emnlp-main.491",
doi = "10.18653/v1/2022.emnlp-main.491",
pages = "7283--7293",
abstract = "Keyphrase generation aims to automatically generate short phrases summarizing an input document. The recently emerged ONE2SET paradigm (Ye et al., 2021) generates keyphrases as a set and has achieved competitive performance. Nevertheless, we observe serious calibration errors outputted by ONE2SET, especially in the over-estimation of ∅ token (means {``}no corresponding keyphrase{''}). In this paper, we deeply analyze this limitation and identify two main reasons behind: 1) the parallel generation has to introduce excessive ∅ as padding tokens into training instances; and 2) the training mechanism assigning target to each slot is unstable and further aggravates the ∅ token over-estimation. To make the model well-calibrated, we propose WR-ONE2SET which extends ONE2SET with an adaptive instance-level cost Weighting strategy and a target Re-assignment mechanism. The former dynamically penalizes the over-estimated slots for different instances thus smoothing the uneven training distribution. The latter refines the original inappropriate assignment and reduces the supervisory signals of over-estimated slots. Experimental results on commonly-used datasets demonstrate the effectiveness and generality of our proposed paradigm.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xie-etal-2022-wr">
<titleInfo>
<title>WR-One2Set: Towards Well-Calibrated Keyphrase Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Binbin</namePart>
<namePart type="family">Xie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiangpeng</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Baosong</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Huan</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Xie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoli</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Min</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jinsong</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Keyphrase generation aims to automatically generate short phrases summarizing an input document. The recently emerged ONE2SET paradigm (Ye et al., 2021) generates keyphrases as a set and has achieved competitive performance. Nevertheless, we observe serious calibration errors outputted by ONE2SET, especially in the over-estimation of ∅ token (means “no corresponding keyphrase”). In this paper, we deeply analyze this limitation and identify two main reasons behind: 1) the parallel generation has to introduce excessive ∅ as padding tokens into training instances; and 2) the training mechanism assigning target to each slot is unstable and further aggravates the ∅ token over-estimation. To make the model well-calibrated, we propose WR-ONE2SET which extends ONE2SET with an adaptive instance-level cost Weighting strategy and a target Re-assignment mechanism. The former dynamically penalizes the over-estimated slots for different instances thus smoothing the uneven training distribution. The latter refines the original inappropriate assignment and reduces the supervisory signals of over-estimated slots. Experimental results on commonly-used datasets demonstrate the effectiveness and generality of our proposed paradigm.</abstract>
<identifier type="citekey">xie-etal-2022-wr</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.491</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.491</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>7283</start>
<end>7293</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T WR-One2Set: Towards Well-Calibrated Keyphrase Generation
%A Xie, Binbin
%A Wei, Xiangpeng
%A Yang, Baosong
%A Lin, Huan
%A Xie, Jun
%A Wang, Xiaoli
%A Zhang, Min
%A Su, Jinsong
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F xie-etal-2022-wr
%X Keyphrase generation aims to automatically generate short phrases summarizing an input document. The recently emerged ONE2SET paradigm (Ye et al., 2021) generates keyphrases as a set and has achieved competitive performance. Nevertheless, we observe serious calibration errors outputted by ONE2SET, especially in the over-estimation of ∅ token (means “no corresponding keyphrase”). In this paper, we deeply analyze this limitation and identify two main reasons behind: 1) the parallel generation has to introduce excessive ∅ as padding tokens into training instances; and 2) the training mechanism assigning target to each slot is unstable and further aggravates the ∅ token over-estimation. To make the model well-calibrated, we propose WR-ONE2SET which extends ONE2SET with an adaptive instance-level cost Weighting strategy and a target Re-assignment mechanism. The former dynamically penalizes the over-estimated slots for different instances thus smoothing the uneven training distribution. The latter refines the original inappropriate assignment and reduces the supervisory signals of over-estimated slots. Experimental results on commonly-used datasets demonstrate the effectiveness and generality of our proposed paradigm.
%R 10.18653/v1/2022.emnlp-main.491
%U https://aclanthology.org/2022.emnlp-main.491
%U https://doi.org/10.18653/v1/2022.emnlp-main.491
%P 7283-7293
Markdown (Informal)
[WR-One2Set: Towards Well-Calibrated Keyphrase Generation](https://aclanthology.org/2022.emnlp-main.491) (Xie et al., EMNLP 2022)
ACL
- Binbin Xie, Xiangpeng Wei, Baosong Yang, Huan Lin, Jun Xie, Xiaoli Wang, Min Zhang, and Jinsong Su. 2022. WR-One2Set: Towards Well-Calibrated Keyphrase Generation. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 7283–7293, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.