Interventional training for out-of-distribution natural language understanding
Out-of-distribution (OOD) settings are used to measure a model’s performance when the distribution of the test data is different from that of the training data. NLU models are known to suffer in OOD settings (Utama et al., 2020b). We study this issue from the perspective of causality, which sees con...
Saved in:
Main Authors: | , , , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2022
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/7548 https://ink.library.smu.edu.sg/context/sis_research/article/8551/viewcontent/Debias_Sicheng.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-8551 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-85512022-11-29T07:07:22Z Interventional training for out-of-distribution natural language understanding YU, Sicheng JIANG, Jing ZHANG, Hao NIU, Yulei SUN, Qianru BING, Lidong Out-of-distribution (OOD) settings are used to measure a model’s performance when the distribution of the test data is different from that of the training data. NLU models are known to suffer in OOD settings (Utama et al., 2020b). We study this issue from the perspective of causality, which sees confounding bias as the reason for models to learn spurious correlations. While a common solution is to perform intervention, existing methods handle only known and single confounder, but in many NLU tasks the confounders can be both unknown and multifactorial. In this paper, we propose a novel interventional training method called Bottom-up Automatic Intervention (BAI) that performs multi-granular intervention with identified multifactorial confounders. Our experiments on three NLU tasks, namely, natural language inference, fact verification and paraphrase identification, show the effectiveness of BAI for tackling OOD settings. 2022-12-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7548 https://ink.library.smu.edu.sg/context/sis_research/article/8551/viewcontent/Debias_Sicheng.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Natural language understanding Out-of-domain detection Dialogue system Text classification Artificial Intelligence and Robotics |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Natural language understanding Out-of-domain detection Dialogue system Text classification Artificial Intelligence and Robotics |
spellingShingle |
Natural language understanding Out-of-domain detection Dialogue system Text classification Artificial Intelligence and Robotics YU, Sicheng JIANG, Jing ZHANG, Hao NIU, Yulei SUN, Qianru BING, Lidong Interventional training for out-of-distribution natural language understanding |
description |
Out-of-distribution (OOD) settings are used to measure a model’s performance when the distribution of the test data is different from that of the training data. NLU models are known to suffer in OOD settings (Utama et al., 2020b). We study this issue from the perspective of causality, which sees confounding bias as the reason for models to learn spurious correlations. While a common solution is to perform intervention, existing methods handle only known and single confounder, but in many NLU tasks the confounders can be both unknown and multifactorial. In this paper, we propose a novel interventional training method called Bottom-up Automatic Intervention (BAI) that performs multi-granular intervention with identified multifactorial confounders. Our experiments on three NLU tasks, namely, natural language inference, fact verification and paraphrase identification, show the effectiveness of BAI for tackling OOD settings. |
format |
text |
author |
YU, Sicheng JIANG, Jing ZHANG, Hao NIU, Yulei SUN, Qianru BING, Lidong |
author_facet |
YU, Sicheng JIANG, Jing ZHANG, Hao NIU, Yulei SUN, Qianru BING, Lidong |
author_sort |
YU, Sicheng |
title |
Interventional training for out-of-distribution natural language understanding |
title_short |
Interventional training for out-of-distribution natural language understanding |
title_full |
Interventional training for out-of-distribution natural language understanding |
title_fullStr |
Interventional training for out-of-distribution natural language understanding |
title_full_unstemmed |
Interventional training for out-of-distribution natural language understanding |
title_sort |
interventional training for out-of-distribution natural language understanding |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2022 |
url |
https://ink.library.smu.edu.sg/sis_research/7548 https://ink.library.smu.edu.sg/context/sis_research/article/8551/viewcontent/Debias_Sicheng.pdf |
_version_ |
1770576370508759040 |