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 |
| Summary: | 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. |
|---|