The Trustworthy Data Science and AI (TDSAI) Lab, also known as Cao Lab, is at the forefront of advancements in trustworthy data science and AI, with a strong emphasis on privacy, security, reliability, fairness, and transparency. Our mission is to develop groundbreaking theories, algorithms, and systems that promote ethical and dependable AI, contributing to the welfare of society and the future of humanity.
Our current research focuses on developing and applying privacy-enhancing technologies to address trust challenges in data management, machine learning, and security. Key areas of exploration include Differential Privacy, Federated Learning, Trustworthy LLM/LMM, AI Security, Data Marketplaces, and beyond.
February 12, 2025
Please consider your submission to ADMA 2025 (Kyoto, October 22-24)! Here is the CFP. Due May 8.
February 9, 2025
Our survey paper on Privacy in Fine-Tuning LLM was accepted in PAKDD 2025. Congrats to all collaborators!
February 6, 2025
Prof. Essam Rashed and Mr.Tamer Ahmed from University of Hyogo gave an intriguing talk titled “Knowledge distillation in medical image segmentation”.
January 24, 2025
Prof. Guandong Xu from EdUHK/UTS gave a distinguished lecture titled “Trustworthy Recommender Systems: Algorithms and Applications”.
January 23, 2025
Our paper on Federated Fine-Tuning was accepted in ICLR 2025. Congrats to all collaborators!
December 25, 2024
Our paper on Federated Graph Condensation was accepted in AAAI 2025. Congrats to all collaborators!
