@inproceedings{li-etal-2022-unified,
title = "A Unified Neural Network Model for Readability Assessment with Feature Projection and Length-Balanced Loss",
author = "Li, Wenbiao and
Ziyang, Wang and
Wu, Yunfang",
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.504",
doi = "10.18653/v1/2022.emnlp-main.504",
pages = "7446--7457",
abstract = "Readability assessment is a basic research task in the field of education. Traditional methods mainly employ machine learning classifiers with hundreds of linguistic features. Although the deep learning model has become the prominent approach for almost all NLP tasks, it is less explored for readability assessment. In this paper, we propose a BERT-based model with feature projection and length-balanced loss (BERT-FP-LBL) to determine the difficulty level of a given text. First, we introduce topic features guided by difficulty knowledge to complement the traditional linguistic features. From the linguistic features, we extract really useful orthogonal features to supplement BERT representations by means of projection filtering. Furthermore, we design a length-balanced loss to handle the greatly varying length distribution of the readability data. We conduct experiments on three English benchmark datasets and one Chinese dataset, and the experimental results show that our proposed model achieves significant improvements over baseline models. Interestingly, our proposed model achieves comparable results with human experts in consistency test.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2022-unified">
<titleInfo>
<title>A Unified Neural Network Model for Readability Assessment with Feature Projection and Length-Balanced Loss</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wenbiao</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wang</namePart>
<namePart type="family">Ziyang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yunfang</namePart>
<namePart type="family">Wu</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>Readability assessment is a basic research task in the field of education. Traditional methods mainly employ machine learning classifiers with hundreds of linguistic features. Although the deep learning model has become the prominent approach for almost all NLP tasks, it is less explored for readability assessment. In this paper, we propose a BERT-based model with feature projection and length-balanced loss (BERT-FP-LBL) to determine the difficulty level of a given text. First, we introduce topic features guided by difficulty knowledge to complement the traditional linguistic features. From the linguistic features, we extract really useful orthogonal features to supplement BERT representations by means of projection filtering. Furthermore, we design a length-balanced loss to handle the greatly varying length distribution of the readability data. We conduct experiments on three English benchmark datasets and one Chinese dataset, and the experimental results show that our proposed model achieves significant improvements over baseline models. Interestingly, our proposed model achieves comparable results with human experts in consistency test.</abstract>
<identifier type="citekey">li-etal-2022-unified</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.504</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.504</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>7446</start>
<end>7457</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Unified Neural Network Model for Readability Assessment with Feature Projection and Length-Balanced Loss
%A Li, Wenbiao
%A Ziyang, Wang
%A Wu, Yunfang
%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 li-etal-2022-unified
%X Readability assessment is a basic research task in the field of education. Traditional methods mainly employ machine learning classifiers with hundreds of linguistic features. Although the deep learning model has become the prominent approach for almost all NLP tasks, it is less explored for readability assessment. In this paper, we propose a BERT-based model with feature projection and length-balanced loss (BERT-FP-LBL) to determine the difficulty level of a given text. First, we introduce topic features guided by difficulty knowledge to complement the traditional linguistic features. From the linguistic features, we extract really useful orthogonal features to supplement BERT representations by means of projection filtering. Furthermore, we design a length-balanced loss to handle the greatly varying length distribution of the readability data. We conduct experiments on three English benchmark datasets and one Chinese dataset, and the experimental results show that our proposed model achieves significant improvements over baseline models. Interestingly, our proposed model achieves comparable results with human experts in consistency test.
%R 10.18653/v1/2022.emnlp-main.504
%U https://aclanthology.org/2022.emnlp-main.504
%U https://doi.org/10.18653/v1/2022.emnlp-main.504
%P 7446-7457
Markdown (Informal)
[A Unified Neural Network Model for Readability Assessment with Feature Projection and Length-Balanced Loss](https://aclanthology.org/2022.emnlp-main.504) (Li et al., EMNLP 2022)
ACL