@inproceedings{wang-etal-2022-feeding,
title = "Feeding What You Need by Understanding What You Learned",
author = "Wang, Xiaoqiang and
Liu, Bang and
Xu, Fangli and
Long, Bo and
Tang, Siliang and
Wu, Lingfei",
editor = "Muresan, Smaranda and
Nakov, Preslav and
Villavicencio, Aline",
booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = may,
year = "2022",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.acl-long.403",
doi = "10.18653/v1/2022.acl-long.403",
pages = "5858--5874",
abstract = "Machine Reading Comprehension (MRC) reveals the ability to understand a given text passage and answer questions based on it. Existing research works in MRC rely heavily on large-size models and corpus to improve the performance evaluated by metrics such as Exact Match ($EM$) and $F_1$. However, such a paradigm lacks sufficient interpretation to model capability and can not efficiently train a model with a large corpus. In this paper, we argue that a deep understanding of model capabilities and data properties can help us feed a model with appropriate training data based on its learning status. Specifically, we design an MRC capability assessment framework that assesses model capabilities in an explainable and multi-dimensional manner. Based on it, we further uncover and disentangle the connections between various data properties and model performance. Finally, to verify the effectiveness of the proposed MRC capability assessment framework, we incorporate it into a curriculum learning pipeline and devise a Capability Boundary Breakthrough Curriculum (CBBC) strategy, which performs a model capability-based training to maximize the data value and improve training efficiency. Extensive experiments demonstrate that our approach significantly improves performance, achieving up to an 11.22{\%} / 8.71{\%} improvement of $EM$ / $F_1$ on MRC tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2022-feeding">
<titleInfo>
<title>Feeding What You Need by Understanding What You Learned</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xiaoqiang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fangli</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bo</namePart>
<namePart type="family">Long</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siliang</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lingfei</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Smaranda</namePart>
<namePart type="family">Muresan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Preslav</namePart>
<namePart type="family">Nakov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aline</namePart>
<namePart type="family">Villavicencio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Machine Reading Comprehension (MRC) reveals the ability to understand a given text passage and answer questions based on it. Existing research works in MRC rely heavily on large-size models and corpus to improve the performance evaluated by metrics such as Exact Match (EM) and F₁. However, such a paradigm lacks sufficient interpretation to model capability and can not efficiently train a model with a large corpus. In this paper, we argue that a deep understanding of model capabilities and data properties can help us feed a model with appropriate training data based on its learning status. Specifically, we design an MRC capability assessment framework that assesses model capabilities in an explainable and multi-dimensional manner. Based on it, we further uncover and disentangle the connections between various data properties and model performance. Finally, to verify the effectiveness of the proposed MRC capability assessment framework, we incorporate it into a curriculum learning pipeline and devise a Capability Boundary Breakthrough Curriculum (CBBC) strategy, which performs a model capability-based training to maximize the data value and improve training efficiency. Extensive experiments demonstrate that our approach significantly improves performance, achieving up to an 11.22% / 8.71% improvement of EM / F₁ on MRC tasks.</abstract>
<identifier type="citekey">wang-etal-2022-feeding</identifier>
<identifier type="doi">10.18653/v1/2022.acl-long.403</identifier>
<location>
<url>https://aclanthology.org/2022.acl-long.403</url>
</location>
<part>
<date>2022-05</date>
<extent unit="page">
<start>5858</start>
<end>5874</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Feeding What You Need by Understanding What You Learned
%A Wang, Xiaoqiang
%A Liu, Bang
%A Xu, Fangli
%A Long, Bo
%A Tang, Siliang
%A Wu, Lingfei
%Y Muresan, Smaranda
%Y Nakov, Preslav
%Y Villavicencio, Aline
%S Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2022
%8 May
%I Association for Computational Linguistics
%C Dublin, Ireland
%F wang-etal-2022-feeding
%X Machine Reading Comprehension (MRC) reveals the ability to understand a given text passage and answer questions based on it. Existing research works in MRC rely heavily on large-size models and corpus to improve the performance evaluated by metrics such as Exact Match (EM) and F₁. However, such a paradigm lacks sufficient interpretation to model capability and can not efficiently train a model with a large corpus. In this paper, we argue that a deep understanding of model capabilities and data properties can help us feed a model with appropriate training data based on its learning status. Specifically, we design an MRC capability assessment framework that assesses model capabilities in an explainable and multi-dimensional manner. Based on it, we further uncover and disentangle the connections between various data properties and model performance. Finally, to verify the effectiveness of the proposed MRC capability assessment framework, we incorporate it into a curriculum learning pipeline and devise a Capability Boundary Breakthrough Curriculum (CBBC) strategy, which performs a model capability-based training to maximize the data value and improve training efficiency. Extensive experiments demonstrate that our approach significantly improves performance, achieving up to an 11.22% / 8.71% improvement of EM / F₁ on MRC tasks.
%R 10.18653/v1/2022.acl-long.403
%U https://aclanthology.org/2022.acl-long.403
%U https://doi.org/10.18653/v1/2022.acl-long.403
%P 5858-5874
Markdown (Informal)
[Feeding What You Need by Understanding What You Learned](https://aclanthology.org/2022.acl-long.403) (Wang et al., ACL 2022)
ACL
- Xiaoqiang Wang, Bang Liu, Fangli Xu, Bo Long, Siliang Tang, and Lingfei Wu. 2022. Feeding What You Need by Understanding What You Learned. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5858–5874, Dublin, Ireland. Association for Computational Linguistics.