@inproceedings{trivedi-etal-2020-multihop,
title = "Is Multihop {QA} in {DiRe} Condition? Measuring and Reducing Disconnected Reasoning",
author = "Trivedi, Harsh and
Balasubramanian, Niranjan and
Khot, Tushar and
Sabharwal, Ashish",
editor = "Webber, Bonnie and
Cohn, Trevor and
He, Yulan and
Liu, Yang",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.emnlp-main.712",
doi = "10.18653/v1/2020.emnlp-main.712",
pages = "8846--8863",
abstract = "Has there been real progress in multi-hop question-answering? Models often exploit dataset artifacts to produce correct answers, without connecting information across multiple supporting facts. This limits our ability to measure true progress and defeats the purpose of building multi-hop QA datasets. We make three contributions towards addressing this. First, we formalize such undesirable behavior as disconnected reasoning across subsets of supporting facts. This allows developing a model-agnostic probe for measuring how much any model can cheat via disconnected reasoning. Second, using a notion of \textit{contrastive support sufficiency}, we introduce an automatic transformation of existing datasets that reduces the amount of disconnected reasoning. Third, our experiments suggest that there hasn{'}t been much progress in multi-hop QA in the reading comprehension setting. For a recent large-scale model (XLNet), we show that only 18 points out of its answer F1 score of 72 on HotpotQA are obtained through multifact reasoning, roughly the same as that of a simpler RNN baseline. Our transformation substantially reduces disconnected reasoning (19 points in answer F1). It is complementary to adversarial approaches, yielding further reductions in conjunction.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="trivedi-etal-2020-multihop">
<titleInfo>
<title>Is Multihop QA in DiRe Condition? Measuring and Reducing Disconnected Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Harsh</namePart>
<namePart type="family">Trivedi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Niranjan</namePart>
<namePart type="family">Balasubramanian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tushar</namePart>
<namePart type="family">Khot</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ashish</namePart>
<namePart type="family">Sabharwal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bonnie</namePart>
<namePart type="family">Webber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Trevor</namePart>
<namePart type="family">Cohn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yulan</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Liu</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>Has there been real progress in multi-hop question-answering? Models often exploit dataset artifacts to produce correct answers, without connecting information across multiple supporting facts. This limits our ability to measure true progress and defeats the purpose of building multi-hop QA datasets. We make three contributions towards addressing this. First, we formalize such undesirable behavior as disconnected reasoning across subsets of supporting facts. This allows developing a model-agnostic probe for measuring how much any model can cheat via disconnected reasoning. Second, using a notion of contrastive support sufficiency, we introduce an automatic transformation of existing datasets that reduces the amount of disconnected reasoning. Third, our experiments suggest that there hasn’t been much progress in multi-hop QA in the reading comprehension setting. For a recent large-scale model (XLNet), we show that only 18 points out of its answer F1 score of 72 on HotpotQA are obtained through multifact reasoning, roughly the same as that of a simpler RNN baseline. Our transformation substantially reduces disconnected reasoning (19 points in answer F1). It is complementary to adversarial approaches, yielding further reductions in conjunction.</abstract>
<identifier type="citekey">trivedi-etal-2020-multihop</identifier>
<identifier type="doi">10.18653/v1/2020.emnlp-main.712</identifier>
<location>
<url>https://aclanthology.org/2020.emnlp-main.712</url>
</location>
<part>
<date>2020-11</date>
<extent unit="page">
<start>8846</start>
<end>8863</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Is Multihop QA in DiRe Condition? Measuring and Reducing Disconnected Reasoning
%A Trivedi, Harsh
%A Balasubramanian, Niranjan
%A Khot, Tushar
%A Sabharwal, Ashish
%Y Webber, Bonnie
%Y Cohn, Trevor
%Y He, Yulan
%Y Liu, Yang
%S Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
%D 2020
%8 November
%I Association for Computational Linguistics
%C Online
%F trivedi-etal-2020-multihop
%X Has there been real progress in multi-hop question-answering? Models often exploit dataset artifacts to produce correct answers, without connecting information across multiple supporting facts. This limits our ability to measure true progress and defeats the purpose of building multi-hop QA datasets. We make three contributions towards addressing this. First, we formalize such undesirable behavior as disconnected reasoning across subsets of supporting facts. This allows developing a model-agnostic probe for measuring how much any model can cheat via disconnected reasoning. Second, using a notion of contrastive support sufficiency, we introduce an automatic transformation of existing datasets that reduces the amount of disconnected reasoning. Third, our experiments suggest that there hasn’t been much progress in multi-hop QA in the reading comprehension setting. For a recent large-scale model (XLNet), we show that only 18 points out of its answer F1 score of 72 on HotpotQA are obtained through multifact reasoning, roughly the same as that of a simpler RNN baseline. Our transformation substantially reduces disconnected reasoning (19 points in answer F1). It is complementary to adversarial approaches, yielding further reductions in conjunction.
%R 10.18653/v1/2020.emnlp-main.712
%U https://aclanthology.org/2020.emnlp-main.712
%U https://doi.org/10.18653/v1/2020.emnlp-main.712
%P 8846-8863
Markdown (Informal)
[Is Multihop QA in DiRe Condition? Measuring and Reducing Disconnected Reasoning](https://aclanthology.org/2020.emnlp-main.712) (Trivedi et al., EMNLP 2020)
ACL