Robust Universal Adversarial Perturbations
Changming Xu, Gagandeep Singh. “Robust Universal Adversarial Perturbations.” In Submission.
Download here
Publications
*Equal Contribution
Input-Relational Verification of Deep Neural Networks
Debangshu Banerjee, Changming Xu, Gagandeep Singh. “Input-Relational Verification of Deep Neural Networks.” In Submission.
Download here
Bypassing the Safety Training of Open-Source LLMs with Priming Attacks
Jason Vega*, Isha Chaudhary*, Changming Xu*, Gagandeep Singh. “Bypassing the Safety Training of Open-Source LLMs with Priming Attacks.” ICLR Tiny Paper ’24 (invite to present)
Download here
Exploring Practical Vulnerabilities of Machine Learning-based Wireless Systems
Liu, Zikun, Changming Xu, Emerson Sie, Gagandeep Singh, and Deepak Vasisht. “Exploring Practical Vulnerabilities of Machine Learning-based Wireless Systems.” In 20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23), pp. 1801-1817. 2023.
Download here
Race Detection and Reachability in Nearly Series-Parallel DAGs
Kunal Agrawal, Joseph Devietti, Jeremy T. Fineman, I-Ting Angelina Lee, Robert Utterback, and Changming Xu. “Race detection and reachability in nearly series-parallel dags.” ACM-SIAM SODA ‘18.
Download here