@inproceedings{ni-etal-2022-r,
title = "{R}-{AT}: Regularized Adversarial Training for Natural Language Understanding",
author = "Ni, Shiwen and
Li, Jiawen and
Kao, Hung-Yu",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.480",
doi = "10.18653/v1/2022.findings-emnlp.480",
pages = "6427--6440",
abstract = "Currently, adversarial training has become a popular and powerful regularization method in the natural language domain. In this paper, we Regularized Adversarial Training (R-AT) via dropout, which forces the output probability distributions of different sub-models generated by dropout to be consistent under the same adversarial samples. Specifically, we generate adversarial samples by perturbing the word embeddings. For each adversarial sample fed to the model, R-AT minimizes both the adversarial risk and the bidirectional KL-divergence between the adversarial output distributions of two sub-models sampled by dropout. Through extensive experiments on 13 public natural language understanding datasets, we found that R-AT has improvements for many models (e.g., rnn-based, cnn-based, and transformer-based models). For the GLUE benchmark, when R-AT is only applied to the fine-tuning stage, it is able to improve the overall test score of the BERT-base model from 78.3 to 79.6 and the RoBERTa-large model from 88.1 to 88.6. Theoretical analysis reveals that R-AT has potential gradient regularization during the training process. Furthermore, R-AT can reduce the inconsistency between training and testing of models with dropout.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ni-etal-2022-r">
<titleInfo>
<title>R-AT: Regularized Adversarial Training for Natural Language Understanding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shiwen</namePart>
<namePart type="family">Ni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiawen</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hung-Yu</namePart>
<namePart type="family">Kao</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>Findings of the Association for Computational Linguistics: EMNLP 2022</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>Currently, adversarial training has become a popular and powerful regularization method in the natural language domain. In this paper, we Regularized Adversarial Training (R-AT) via dropout, which forces the output probability distributions of different sub-models generated by dropout to be consistent under the same adversarial samples. Specifically, we generate adversarial samples by perturbing the word embeddings. For each adversarial sample fed to the model, R-AT minimizes both the adversarial risk and the bidirectional KL-divergence between the adversarial output distributions of two sub-models sampled by dropout. Through extensive experiments on 13 public natural language understanding datasets, we found that R-AT has improvements for many models (e.g., rnn-based, cnn-based, and transformer-based models). For the GLUE benchmark, when R-AT is only applied to the fine-tuning stage, it is able to improve the overall test score of the BERT-base model from 78.3 to 79.6 and the RoBERTa-large model from 88.1 to 88.6. Theoretical analysis reveals that R-AT has potential gradient regularization during the training process. Furthermore, R-AT can reduce the inconsistency between training and testing of models with dropout.</abstract>
<identifier type="citekey">ni-etal-2022-r</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.480</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.480</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>6427</start>
<end>6440</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T R-AT: Regularized Adversarial Training for Natural Language Understanding
%A Ni, Shiwen
%A Li, Jiawen
%A Kao, Hung-Yu
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F ni-etal-2022-r
%X Currently, adversarial training has become a popular and powerful regularization method in the natural language domain. In this paper, we Regularized Adversarial Training (R-AT) via dropout, which forces the output probability distributions of different sub-models generated by dropout to be consistent under the same adversarial samples. Specifically, we generate adversarial samples by perturbing the word embeddings. For each adversarial sample fed to the model, R-AT minimizes both the adversarial risk and the bidirectional KL-divergence between the adversarial output distributions of two sub-models sampled by dropout. Through extensive experiments on 13 public natural language understanding datasets, we found that R-AT has improvements for many models (e.g., rnn-based, cnn-based, and transformer-based models). For the GLUE benchmark, when R-AT is only applied to the fine-tuning stage, it is able to improve the overall test score of the BERT-base model from 78.3 to 79.6 and the RoBERTa-large model from 88.1 to 88.6. Theoretical analysis reveals that R-AT has potential gradient regularization during the training process. Furthermore, R-AT can reduce the inconsistency between training and testing of models with dropout.
%R 10.18653/v1/2022.findings-emnlp.480
%U https://aclanthology.org/2022.findings-emnlp.480
%U https://doi.org/10.18653/v1/2022.findings-emnlp.480
%P 6427-6440
Markdown (Informal)
[R-AT: Regularized Adversarial Training for Natural Language Understanding](https://aclanthology.org/2022.findings-emnlp.480) (Ni et al., Findings 2022)
ACL