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. For instance, backdoors can be implanted through crafting training instances with a specific textual trigger and a target label. This paper posits that backdoor poisoning attacks exhibit \emph{spurious correlation} between simple text features and classification labels, and accordingly, proposes methods for mitigating spurious correlation as means of defence. Our empirical study reveals that the malicious triggers are highly correlated to their target labels; therefore such correlations are extremely distinguishable compared to those scores of benign features, and can be used to filter out potentially problematic instances. Compared with several existing defences, our defence method significantly reduces attack success rates across backdoor attacks, and in the case of insertion-based attacks, our method provides a near-perfect defence.
Code (1)
Similar Papers 제목 키워드 기반
Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning
Recently, various parameter-efficient fine-tuning (PEFT) strategies for application to language models have been proposed and successfully implemented. However, this raises the question of whether PEFT, which only update…
Backdoor Attackparameter-efficient fine-tuningtext-classificationText ClassificationPoisHygiene: Detecting and Mitigating Poisoning Attacks in Neural Networks
The black-box nature of deep neural networks (DNNs) facilitates attackers to manipulate the behavior of DNN through data poisoning. Being able to detect and mitigate poisoning attacks, typically categorized into backdoor…
Data PoisoningMitigating backdoor attacks in LSTM-based Text Classification Systems by Backdoor Keyword Identification
It has been proved that deep neural networks are facing a new threat called backdoor attacks, where the adversary can inject backdoors into the neural network model through poisoning the training dataset. When the input …
ClassificationData PoisoningGeneral Classificationtext-classification+1Mitigating Data Poisoning in Text Classification with Differential Privacy
NLP models are vulnerable to data poisoning attacks. One type of attack can plant a backdoor in a model by injecting poisoned examples in training, causing the victim model to misclassify test instances which include a s…
ClassificationData Poisoningtext-classificationText ClassificationSEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
Modern NLP models are often trained on public datasets drawn from diverse sources, rendering them vulnerable to data poisoning attacks. These attacks can manipulate the model's behavior in ways engineered by the attacker…
Data Poisoning