Downloads: 16

Files in This Item:
File Description SizeFormat 
ACCESS.2023.3308916.pdf1.72 MBAdobe PDFView/Open
Title: Japanese Event Factuality Analysis in the Era of BERT
Authors: Kameko, Hirotaka
Murawaki, Yugo  kyouindb  KAKEN_id  orcid (unconfirmed)
Matsuyoshi, Suguru
Mori, Shinsuke
Author's alias: 亀甲, 博貴
村脇, 有吾
森, 信介
Keywords: Task analysis
Training data
Online services
Neural networks
Event detection
Sequential analysis
Event factuality
sequence labeling
neural networks
multi-task learning
Issue Date: 2023
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Journal title: IEEE Access
Volume: 11
Start page: 93286
End page: 93292
Abstract: Recognizing event factuality is a crucial factor for understanding and generating texts with abundant references to possible and counterfactual events. Because event factuality is signaled by modality expressions, identifying modality expression is also an important task. The question then is how to solve these interconnected tasks. On the one hand, while neural networks facilitate multi-task learning by means of parameter sharing among related tasks, the recently introduced pre-training/fine-tuning paradigm might be powerful enough for the model to be able to learn one task without indirect signals from another. On the other hand, ever-increasing model sizes make it practically difficult to run multiple task-specific fine-tuned models at inference time so that parameter sharing can be seen as an effective way to reduce the model’s size. Through experiments, we found: (1) BERT-CRF outperformed non-neural models and BiLSTM-CRF; (2) BERT-CRF did neither benefit from nor was negatively impacted by multi-task learning, indicating the practical viability of BERT-CRF combined with multi-task learning.
Rights: This work is licensed under a Creative Commons Attribution 4.0 License.
DOI(Published Version): 10.1109/ACCESS.2023.3308916
Appears in Collections:Journal Articles

Show full item record

Export to RefWorks

Export Format: 

This item is licensed under a Creative Commons License Creative Commons