arXiv · 2010.02684
Poison Attacks against Text Datasets with Conditional Adversarially Regularized Autoencoder
Abstract
This paper demonstrates a fatal vulnerability in natural language inference (NLI) and text classification systems. More concretely, we present a 'backdoor poisoning' attack on NLP models. Our poisoning attack utilizes conditional adversarially regularized autoencoder (CARA) to generate poisoned training samples by poison injection in latent space. Just by adding 1% poisoned data, our experiments show that a victim BERT finetuned classifier's predictions can be steered to the poison target class with success rates of >80% when the input hypothesis is injected with the poison signature, demonstrating that NLI and text classification systems face a huge security risk.
Explore related subjects
Keep this discovery
Alvin Chan, Yi Tay, Yew-Soon Ong, Aston Zhang. 2020-10-06. Poison Attacks against Text Datasets with Conditional Adversarially Regularized Autoencoder. https://arxiv.org/abs/2010.02684
Cite the original work for its findings. Save a collection to share your selection of sources.