From Dataset Recycling to Multi-Property Extraction and Beyond

Tytuł:
From Dataset Recycling to Multi-Property Extraction and Beyond
Konferencja:
Conference on Computational Natural Language Learning [CoNLL]
Rok:
2020

Opis:
This paper investigates various Transformer architectures on the WikiReading Information Extraction and Machine Reading Comprehension dataset. The proposed Dual-source model outperforms the current state-of-the-art by a large margin. Next, we introduce WikiReading Recycled - a newly developed public dataset, and the task of multiple-property extraction.

Tom (seria wydawnicza):
Proceedings of the 24th Conference on Computational Natural Language Learning

Link:
https://www.aclweb.org/anthology/2020.conll-1.52/

DBLP link:
https://dblp.org/rec/journals/corr/abs-2006-08281.html