Evaluating and Enhancing the Robustness of Sustainable Neural Relationship Classifiers Using Query-E...
Evaluating and Enhancing the Robustness of Sustainable Neural Relationship Classifiers Using Query-Efficient Black-Box Adversarial Attacks
About this item
Full title
Author / Creator
Publisher
Basel: MDPI AG
Journal title
Language
English
Formats
Publication information
Publisher
Basel: MDPI AG
Subjects
More information
Scope and Contents
Contents
Neural relation extraction (NRE) models are the backbone of various machine learning tasks, including knowledge base enrichment, information extraction, and document summarization. Despite the vast popularity of these models, their vulnerabilities remain unknown; this is of high concern given their growing use in security-sensitive applications suc...
Alternative Titles
Full title
Evaluating and Enhancing the Robustness of Sustainable Neural Relationship Classifiers Using Query-Efficient Black-Box Adversarial Attacks
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_2539991783
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2539991783
Other Identifiers
ISSN
2071-1050
E-ISSN
2071-1050
DOI
10.3390/su13115892
