@inproceedings{swanson-etal-2020-rationalizing,
title = "Rationalizing Text Matching: {L}earning Sparse Alignments via Optimal Transport",
author = "Swanson, Kyle and
Yu, Lili and
Lei, Tao",
editor = "Jurafsky, Dan and
Chai, Joyce and
Schluter, Natalie and
Tetreault, Joel",
booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
month = jul,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.acl-main.496",
doi = "10.18653/v1/2020.acl-main.496",
pages = "5609--5626",
abstract = "Selecting input features of top relevance has become a popular method for building self-explaining models. In this work, we extend this selective rationalization approach to text matching, where the goal is to jointly select and align text pieces, such as tokens or sentences, as a justification for the downstream prediction. Our approach employs optimal transport (OT) to find a minimal cost alignment between the inputs. However, directly applying OT often produces dense and therefore uninterpretable alignments. To overcome this limitation, we introduce novel constrained variants of the OT problem that result in highly sparse alignments with controllable sparsity. Our model is end-to-end differentiable using the Sinkhorn algorithm for OT and can be trained without any alignment annotations. We evaluate our model on the StackExchange, MultiNews, e-SNLI, and MultiRC datasets. Our model achieves very sparse rationale selections with high fidelity while preserving prediction accuracy compared to strong attention baseline models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="swanson-etal-2020-rationalizing">
<titleInfo>
<title>Rationalizing Text Matching: Learning Sparse Alignments via Optimal Transport</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kyle</namePart>
<namePart type="family">Swanson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lili</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tao</namePart>
<namePart type="family">Lei</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 58th Annual Meeting of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dan</namePart>
<namePart type="family">Jurafsky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Chai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Natalie</namePart>
<namePart type="family">Schluter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joel</namePart>
<namePart type="family">Tetreault</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>Selecting input features of top relevance has become a popular method for building self-explaining models. In this work, we extend this selective rationalization approach to text matching, where the goal is to jointly select and align text pieces, such as tokens or sentences, as a justification for the downstream prediction. Our approach employs optimal transport (OT) to find a minimal cost alignment between the inputs. However, directly applying OT often produces dense and therefore uninterpretable alignments. To overcome this limitation, we introduce novel constrained variants of the OT problem that result in highly sparse alignments with controllable sparsity. Our model is end-to-end differentiable using the Sinkhorn algorithm for OT and can be trained without any alignment annotations. We evaluate our model on the StackExchange, MultiNews, e-SNLI, and MultiRC datasets. Our model achieves very sparse rationale selections with high fidelity while preserving prediction accuracy compared to strong attention baseline models.</abstract>
<identifier type="citekey">swanson-etal-2020-rationalizing</identifier>
<identifier type="doi">10.18653/v1/2020.acl-main.496</identifier>
<location>
<url>https://aclanthology.org/2020.acl-main.496</url>
</location>
<part>
<date>2020-07</date>
<extent unit="page">
<start>5609</start>
<end>5626</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Rationalizing Text Matching: Learning Sparse Alignments via Optimal Transport
%A Swanson, Kyle
%A Yu, Lili
%A Lei, Tao
%Y Jurafsky, Dan
%Y Chai, Joyce
%Y Schluter, Natalie
%Y Tetreault, Joel
%S Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
%D 2020
%8 July
%I Association for Computational Linguistics
%C Online
%F swanson-etal-2020-rationalizing
%X Selecting input features of top relevance has become a popular method for building self-explaining models. In this work, we extend this selective rationalization approach to text matching, where the goal is to jointly select and align text pieces, such as tokens or sentences, as a justification for the downstream prediction. Our approach employs optimal transport (OT) to find a minimal cost alignment between the inputs. However, directly applying OT often produces dense and therefore uninterpretable alignments. To overcome this limitation, we introduce novel constrained variants of the OT problem that result in highly sparse alignments with controllable sparsity. Our model is end-to-end differentiable using the Sinkhorn algorithm for OT and can be trained without any alignment annotations. We evaluate our model on the StackExchange, MultiNews, e-SNLI, and MultiRC datasets. Our model achieves very sparse rationale selections with high fidelity while preserving prediction accuracy compared to strong attention baseline models.
%R 10.18653/v1/2020.acl-main.496
%U https://aclanthology.org/2020.acl-main.496
%U https://doi.org/10.18653/v1/2020.acl-main.496
%P 5609-5626
Markdown (Informal)
[Rationalizing Text Matching: Learning Sparse Alignments via Optimal Transport](https://aclanthology.org/2020.acl-main.496) (Swanson et al., ACL 2020)
ACL