Defending Against Stealthy Backdoor Attacks
Sangeet Sagar, Abhinav Bhatt, Abhijith Srinivas Bidaralli

TL;DR
This paper explores defense strategies against stealthy backdoor attacks in NLP models, demonstrating methods that reduce attack success while preserving model performance and efficiency.
Contribution
The paper introduces new defense techniques against backdoor attacks that effectively detect and mitigate malicious triggers with minimal impact on benign inputs.
Findings
Defense methods significantly reduce attack success rate.
Maintains high accuracy on benign inputs.
Some defenses are computationally efficient.
Abstract
Defenses against security threats have been an interest of recent studies. Recent works have shown that it is not difficult to attack a natural language processing (NLP) model while defending against them is still a cat-mouse game. Backdoor attacks are one such attack where a neural network is made to perform in a certain way on specific samples containing some triggers while achieving normal results on other samples. In this work, we present a few defense strategies that can be useful to counter against such an attack. We show that our defense methodologies significantly decrease the performance on the attacked inputs while maintaining similar performance on benign inputs. We also show that some of our defenses have very less runtime and also maintain similarity with the original inputs.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdversarial Robustness in Machine Learning · Network Security and Intrusion Detection · Advanced Malware Detection Techniques
