Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation

Modern NLP models are often trained over large untrusted datasets, raising the potential for a malicious adversary to compromise model behaviour.

Data and Resources

Cite this as

Xuanli He, Qiongkai Xu, Jun Wang, Benjamin Rubinstein, Trevor Cohn (2025). Dataset: Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation. https://doi.org/10.57702/3w4etk2g

DOI retrieved: January 2, 2025

Additional Info

Field Value
Created January 2, 2025
Last update January 2, 2025
Defined In https://doi.org/10.48550/arXiv.2305.11596
Author Xuanli He
More Authors
Qiongkai Xu
Jun Wang
Benjamin Rubinstein
Trevor Cohn
Homepage https://github.com/xlhex/emnlp2023_z-defence.git