Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation

Joe Stacey, Marek Rei


Abstract
Knowledge distillation optimises a smaller student model to behave similarly to a larger teacher model, retaining some of the performance benefits. While this method can improve results on in-distribution examples, it does not necessarily generalise to out-of-distribution (OOD) settings. We investigate two complementary methods for improving the robustness of the resulting student models on OOD domains. The first approach augments the distillation with generated unlabeled examples that match the target distribution. The second method upsamples data points among the training set that are similar to the target distribution. When applied on the task of natural language inference (NLI), our experiments on MNLI show that distillation with these modifications outperforms previous robustness solutions. We also find that these methods improve performance on OOD domains even beyond the target domain.
Anthology ID:
2024.findings-acl.132
Volume:
Findings of the Association for Computational Linguistics: ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2239–2258
Language:
URL:
https://aclanthology.org/2024.findings-acl.132
DOI:
10.18653/v1/2024.findings-acl.132
Bibkey:
Cite (ACL):
Joe Stacey and Marek Rei. 2024. Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation. In Findings of the Association for Computational Linguistics: ACL 2024, pages 2239–2258, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation (Stacey & Rei, Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-acl.132.pdf