Tu Ouyang (欧阳图)
- My work and interests focus on enabling meaningful services via effective, efficient, and trustworthy automation and intelligence
- My Google Scholar profile
- My professional profile
- Sep 2024: A poster was presented at Stanford Trust and Safety Research Conference 2024: Beyond the Regular Benchmarks: Evaluate Large Foundation Models’ Potential Usage in Adversarial Activities
- Jan 2024: A new approach of membership inference attack against ML models, work in collaboration with Prof. An Wang's lab at Case Western Reserve University, accepted by IEEE Euro Symposium on Security and Privacy 2024. arXiv version: Learning-Based Difficulty Calibration for Enhanced Membership Inference Attacks
- Apr 2023: A blog post on a method of leveraging patent data for prediction of industrial evolvement: 基于海量专利数据对产业发展进行预测的算法对比
- Aug 2022: A blog post on performance evaluation of using an OLAP (Databricks Delta lake in this case) as an OLTP: 服务后端通过ODBC连接Databricks(on Azure)
- Mar 2022: A paper on principles and practical considerations of migrating apps to the cloud, in collaboration with the University of Melbourne and Zilian Tech, accepted by IECIS 2022: Lifting Existing Applications to the Cloud: Abstractions, Separation of Responsibilities and Tooling Support
- Dec 2021: A paper in collaboration with UESTC and Zilian Tech, accepted by Springer magazine: Leveraging Modern Big Data Stack for Swift Development of Insights into Social Developments