<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2022-unbiased">
    <titleInfo>
        <title>Unbiased Math Word Problems Benchmark for Mitigating Solving Bias</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zhicheng</namePart>
        <namePart type="family">Yang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jinghui</namePart>
        <namePart type="family">Qin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiaqi</namePart>
        <namePart type="family">Chen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xiaodan</namePart>
        <namePart type="family">Liang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2022-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: NAACL 2022</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Marine</namePart>
            <namePart type="family">Carpuat</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Marie-Catherine</namePart>
            <namePart type="family">de Marneffe</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ivan</namePart>
            <namePart type="given">Vladimir</namePart>
            <namePart type="family">Meza Ruiz</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Seattle, United States</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>In this paper, we revisit the solving bias when evaluating models on current Math Word Problem (MWP) benchmarks. However, current solvers exist solving bias which consists of data bias and learning bias due to biased dataset and improper training strategy. Our experiments verify MWP solvers are easy to be biased by the biased training datasets which do not cover diverse questions for each problem narrative of all MWPs, thus a solver can only learn shallow heuristics rather than deep semantics for understanding problems. Besides, an MWP can be naturally solved by multiple equivalent equations while current datasets take only one of the equivalent equations as ground truth, forcing the model to match the labeled ground truth and ignoring other equivalent equations. Here, we first introduce a novel MWP dataset named UnbiasedMWP which is constructed by varying the grounded expressions in our collected data and annotating them with corresponding multiple new questions manually. Then, to further mitigate learning bias, we propose a Dynamic Target Selection (DTS) Strategy to dynamically select more suitable target expressions according to the longest prefix match between the current model output and candidate equivalent equations which are obtained by applying commutative law during training. The results show that our UnbiasedMWP has significantly fewer biases than its original data and other datasets, posing a promising benchmark for fairly evaluating the solvers’ reasoning skills rather than matching nearest neighbors. And the solvers trained with our DTS achieve higher accuracies on multiple MWP benchmarks. The source code is available at https://github.com/yangzhch6/UnbiasedMWP.</abstract>
    <identifier type="citekey">yang-etal-2022-unbiased</identifier>
    <identifier type="doi">10.18653/v1/2022.findings-naacl.104</identifier>
    <location>
        <url>https://aclanthology.org/2022.findings-naacl.104</url>
    </location>
    <part>
        <date>2022-07</date>
        <extent unit="page">
            <start>1401</start>
            <end>1408</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Unbiased Math Word Problems Benchmark for Mitigating Solving Bias
%A Yang, Zhicheng
%A Qin, Jinghui
%A Chen, Jiaqi
%A Liang, Xiaodan
%Y Carpuat, Marine
%Y de Marneffe, Marie-Catherine
%Y Meza Ruiz, Ivan Vladimir
%S Findings of the Association for Computational Linguistics: NAACL 2022
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, United States
%F yang-etal-2022-unbiased
%X In this paper, we revisit the solving bias when evaluating models on current Math Word Problem (MWP) benchmarks. However, current solvers exist solving bias which consists of data bias and learning bias due to biased dataset and improper training strategy. Our experiments verify MWP solvers are easy to be biased by the biased training datasets which do not cover diverse questions for each problem narrative of all MWPs, thus a solver can only learn shallow heuristics rather than deep semantics for understanding problems. Besides, an MWP can be naturally solved by multiple equivalent equations while current datasets take only one of the equivalent equations as ground truth, forcing the model to match the labeled ground truth and ignoring other equivalent equations. Here, we first introduce a novel MWP dataset named UnbiasedMWP which is constructed by varying the grounded expressions in our collected data and annotating them with corresponding multiple new questions manually. Then, to further mitigate learning bias, we propose a Dynamic Target Selection (DTS) Strategy to dynamically select more suitable target expressions according to the longest prefix match between the current model output and candidate equivalent equations which are obtained by applying commutative law during training. The results show that our UnbiasedMWP has significantly fewer biases than its original data and other datasets, posing a promising benchmark for fairly evaluating the solvers’ reasoning skills rather than matching nearest neighbors. And the solvers trained with our DTS achieve higher accuracies on multiple MWP benchmarks. The source code is available at https://github.com/yangzhch6/UnbiasedMWP.
%R 10.18653/v1/2022.findings-naacl.104
%U https://aclanthology.org/2022.findings-naacl.104
%U https://doi.org/10.18653/v1/2022.findings-naacl.104
%P 1401-1408

Markdown (Informal)

[Unbiased Math Word Problems Benchmark for Mitigating Solving Bias](https://aclanthology.org/2022.findings-naacl.104) (Yang et al., Findings 2022)

• Unbiased Math Word Problems Benchmark for Mitigating Solving Bias (Yang et al., Findings 2022)

ACL

• Zhicheng Yang, Jinghui Qin, Jiaqi Chen, and Xiaodan Liang. 2022. Unbiased Math Word Problems Benchmark for Mitigating Solving Bias. In Findings of the Association for Computational Linguistics: NAACL 2022, pages 1401–1408, Seattle, United States. Association for Computational Linguistics.