@inproceedings{yang-etal-2022-robust,
title = "Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering",
author = "Yang, Shiquan and
Huang, Xinting and
Lau, Jey Han and
Erfani, Sarah",
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.88",
doi = "10.18653/v1/2022.findings-emnlp.88",
pages = "1220--1234",
abstract = "Data artifacts incentivize machine learning models to learn non-transferable generalizations by taking advantage of shortcuts in the data, andthere is growing evidence that data artifacts play a role for the strong results that deep learning models achieve in recent natural language processing benchmarks.In this paper, we focus on task-oriented dialogue and investigate whether popular datasets such as MultiWOZ contain such data artifacts.We found that by only keeping frequent phrases in the trainingexamples, state-of-the-art models perform similarly compared to the variant trained with full data, suggesting they exploit these spurious correlationsto solve the task. Motivated by this, we propose a contrastive learning based framework to encourage the model to ignore these cues and focus on learning generalisable patterns. We also experiment with adversarial filtering to remove easy training instances so that the model would focus on learning from the harder instances. We conduct a number of generalization experiments {---} e.g., cross-domain/dataset and adversarial tests {---} to assess the robustness of our approach and found that it works exceptionally well.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2022-robust">
<titleInfo>
<title>Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shiquan</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinting</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jey</namePart>
<namePart type="given">Han</namePart>
<namePart type="family">Lau</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sarah</namePart>
<namePart type="family">Erfani</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>Data artifacts incentivize machine learning models to learn non-transferable generalizations by taking advantage of shortcuts in the data, andthere is growing evidence that data artifacts play a role for the strong results that deep learning models achieve in recent natural language processing benchmarks.In this paper, we focus on task-oriented dialogue and investigate whether popular datasets such as MultiWOZ contain such data artifacts.We found that by only keeping frequent phrases in the trainingexamples, state-of-the-art models perform similarly compared to the variant trained with full data, suggesting they exploit these spurious correlationsto solve the task. Motivated by this, we propose a contrastive learning based framework to encourage the model to ignore these cues and focus on learning generalisable patterns. We also experiment with adversarial filtering to remove easy training instances so that the model would focus on learning from the harder instances. We conduct a number of generalization experiments — e.g., cross-domain/dataset and adversarial tests — to assess the robustness of our approach and found that it works exceptionally well.</abstract>
<identifier type="citekey">yang-etal-2022-robust</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.88</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.88</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>1220</start>
<end>1234</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering
%A Yang, Shiquan
%A Huang, Xinting
%A Lau, Jey Han
%A Erfani, Sarah
%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 yang-etal-2022-robust
%X Data artifacts incentivize machine learning models to learn non-transferable generalizations by taking advantage of shortcuts in the data, andthere is growing evidence that data artifacts play a role for the strong results that deep learning models achieve in recent natural language processing benchmarks.In this paper, we focus on task-oriented dialogue and investigate whether popular datasets such as MultiWOZ contain such data artifacts.We found that by only keeping frequent phrases in the trainingexamples, state-of-the-art models perform similarly compared to the variant trained with full data, suggesting they exploit these spurious correlationsto solve the task. Motivated by this, we propose a contrastive learning based framework to encourage the model to ignore these cues and focus on learning generalisable patterns. We also experiment with adversarial filtering to remove easy training instances so that the model would focus on learning from the harder instances. We conduct a number of generalization experiments — e.g., cross-domain/dataset and adversarial tests — to assess the robustness of our approach and found that it works exceptionally well.
%R 10.18653/v1/2022.findings-emnlp.88
%U https://aclanthology.org/2022.findings-emnlp.88
%U https://doi.org/10.18653/v1/2022.findings-emnlp.88
%P 1220-1234
Markdown (Informal)
[Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering](https://aclanthology.org/2022.findings-emnlp.88) (Yang et al., Findings 2022)
ACL