Publication: Joint training with semantic role labeling for better generalization in natural language inference
Loading...
Program
KU-Authors
Organization Authors
Co-Authors
Date
Language
Embargo Status
No
Journal Title
Journal ISSN
Volume Title
Alternative Title
Abstract
End-to-end models trained on natural language inference (NLI) datasets show low generalization on out-of-distribution evaluation sets. The models tend to learn shallow heuristics due to dataset biases. The performance decreases dramatically on diagnostic sets measuring compositionality or robustness against simple heuristics. Existing solutions for this problem employ dataset augmentation which has the drawbacks of being applicable to only a limited set of adversaries and at worst hurting the model performance on other adversaries not included in the augmentation set. Our proposed solution is to improve sentence understanding (hence out-of-distribution generalization) with joint learning of explicit semantics. We show that a BERT based model trained jointly on English semantic role labeling (SRL) and NLI achieves significantly higher performance on external evaluation sets measuring generalization performance.
Source
Publisher
Association for Computational Linguistics
Subject
Citation
item.page.haspartof
Source
5th Workshop on Representation Learning for NLP (RepL4NLP 2020)
item.page.ispartofseries
item.page.edition
DOI
item.page.datauri
item.page.link
Rights
Other
Copyrights Note
Rights and licensing
Other
Collections
Endorsement
Review
Supplemented By
Referenced By
Google Scholar
Scholar'da Ara ↗ Bu yayında DOI yok — Altmetric/Dimensions/PlumX/BIP! rozetleri DOI gerektirir.
