A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
MCScript2.0: A Machine Comprehension Corpus Focused on Script Events and Participants
2019
Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM 2019)
Preserved Fulltext
We introduce MCScript2.0, a machine comprehension corpus for the end-to-end evaluation of script knowledge. ...
We give a thorough analysis of our corpus and show that while the task is not challenging to humans, existing machine comprehension models fail to perform well on the data, even if they make use of a commonsense ...
We also thank the numerous workers on MTurk for their good work and Carina Silberer and the reviewers for their helpful comments on the paper. ...
doi:10.18653/v1/s19-1012
dblp:conf/starsem/OstermannRP19
fatcat:jmkc4lebjzfo7fqrqxc4yxpwfm
English Machine Reading Comprehension Datasets: A Survey
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
This paper surveys 60 English Machine Reading Comprehension datasets, with a view to providing a convenient resource for other researchers interested in this problem. ...
Our analysis reveals that Wikipedia is by far the most common data source and that there is a relative lack of why, when, and where questions across datasets. ...
We also thank Andrew Dunne, Koel Dutta Chowdhury, Valeriia Filimonova, Victoria Serga, Marina Lisuk, Ke Hu, Joachim Wagner, and Alberto Poncelas. ...
arXiv:2101.10421v2
fatcat:xdkiczo3zzdclgbwpxgamvtgwm
Commonsense Knowledge in Word Associations and ConceptNet
2021
Proceedings of the 25th Conference on Computational Natural Language Learning
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
RN with ConceptNet on OBQA where p=0.07. 13 See Appendix A for details of 17 and 7 relation types. ...
This paper presents an in-depth comparison of two large-scale resources of general knowledge: ConceptNet, an engineered relational database, and SWOW a knowledge graph derived from crowd-sourced word associations ...
We thank the reviewers for their valuable comments, and Simon De Deyne for insightful discussions. ...
doi:10.18653/v1/2021.conll-1.38
fatcat:yxwxo5fiova25lg4qykxzkf4sm
Identify, Align, and Integrate: Matching Knowledge Graphs to Commonsense Reasoning Tasks
2021
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
We present an approach to assess how well a candidate KG can correctly identify and accurately fill in gaps of reasoning for a task, which we call KG-to-task match. ...
We show this KGto-task match in 3 phases: knowledge-task identification, knowledge-task alignment, and knowledge-task integration. ...
This work was supported by DARPA MCS Grant #N66001-19-2-4031, NSF-CAREER Award 1846185, NSF PhD Fellowship, and awards from Microsoft and Amazon. ...
doi:10.18653/v1/2021.eacl-main.192
fatcat:srlgycbh3rajlkgvf2fr5jicqq
Relation-aware Bidirectional Path Reasoning for Commonsense Question Answering
2021
Proceedings of the 25th Conference on Computational Natural Language Learning
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Mcscript2.0: A machine comprehension cor- Haocheng Wu, Zuohui Tian, Wei Wu, and Enhong
pus focused on script events and participants. pages Chen. 2017. ...
The above models are pre- and 72.3% with ELECTRA-large. This indicates
trained on a large text corpus and then fine tuned on that the concept representations obtained from the
the training data. ...
doi:10.18653/v1/2021.conll-1.35
fatcat:b3ev34ntx5dxta5m7lpwfmc2ci