Single Word Replacement Attacks

The dataset is created by modifying SNLI examples with single word replacements that test lexical inferences and simple world knowledge, aimed at analyzing model robustness against simplistic lexical relations.

Data and Resources

Cite this as

M. Glockner, V. Shwartz, Y. Goldberg (2024). Dataset: Single Word Replacement Attacks. https://doi.org/10.57702/ibevyl1r

DOI retrieved: November 25, 2024

Additional Info

Field Value
Created November 25, 2024
Last update November 25, 2024
Defined In https://doi.org/10.48550/arXiv.1912.03441
Author M. Glockner
More Authors
V. Shwartz
Y. Goldberg