4 papers
Double Backdoored: Converting Code Large Language Model Backdoors to Traditional Malware via Adversarial Instruction Tuning Attacks
Md Imran Hossen, Sai Venkatesh Chilukoti, Liqun Shan +3
Instruction-tuned Large Language Models designed for coding tasks are increasingly employed as AI coding assistants. However, the cybersecurity vulnerabilities and implications ari…
Unified Kernel-Segregated Transpose Convolution Operation
Vijay Srinivas Tida, Md Imran Hossen, Liqun Shan +3
The optimization of the transpose convolution layer for deep learning applications is achieved with the kernel segregation mechanism. However, kernel segregation has disadvantages,…
Differentially private fine-tuned NF-Net to predict GI cancer type
Sai Venkatesh Chilukoti, Imran Hossen Md, Liqun Shan +2
Based on global genomic status, the cancer tumor is classified as Microsatellite Instable (MSI) and Microsatellite Stable (MSS). Immunotherapy is used to diagnose MSI, whereas radi…
DP-SGD-Global-Adapt-V2-S: Triad Improvements of Privacy, Accuracy and Fairness via Step Decay Noise Multiplier and Step Decay Upper Clipping Threshold
Sai Venkatesh Chilukoti, Md Imran Hossen, Liqun Shan +4
Differentially Private Stochastic Gradient Descent (DP-SGD) has become a widely used technique for safeguarding sensitive information in deep learning applications. Unfortunately,…