KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments

Ameer Saadat-Yazdi, Xue Li, Sandrine Chausson, Vaishak Belle, Björn Ross, Jeff Z. Pan, Nadin Kökciyan


Abstract
The ArgMining 2022 Shared Task is concerned with predicting the validity and novelty of an inference for a given premise and conclusion pair. We propose two feed-forward network based models (KEViN1 and KEViN2), which combine features generated from several pretrained transformers and the WikiData knowledge graph. The transformers are used to predict entailment and semantic similarity, while WikiData is used to provide a semantic measure between concepts in the premise-conclusion pair. Our proposed models show significant improvement over RoBERTa, with KEViN1 outperforming KEViN2 and obtaining second rank on both subtasks (A and B) of the ArgMining 2022 Shared Task.
Anthology ID:
2022.argmining-1.9
Volume:
Proceedings of the 9th Workshop on Argument Mining
Month:
October
Year:
2022
Address:
Online and in Gyeongju, Republic of Korea
Editors:
Gabriella Lapesa, Jodi Schneider, Yohan Jo, Sougata Saha
Venue:
ArgMining
SIG:
Publisher:
International Conference on Computational Linguistics
Note:
Pages:
104–110
Language:
URL:
https://aclanthology.org/2022.argmining-1.9
DOI:
Bibkey:
Cite (ACL):
Ameer Saadat-Yazdi, Xue Li, Sandrine Chausson, Vaishak Belle, Björn Ross, Jeff Z. Pan, and Nadin Kökciyan. 2022. KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments. In Proceedings of the 9th Workshop on Argument Mining, pages 104–110, Online and in Gyeongju, Republic of Korea. International Conference on Computational Linguistics.
Cite (Informal):
KEViN: A Knowledge Enhanced Validity and Novelty Classifier for Arguments (Saadat-Yazdi et al., ArgMining 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.argmining-1.9.pdf
Data
ValNov Subtask A