Speaker

Yao Qiang, Wayne State University

Abstract

As our society moves increasingly towards being AI-centric, the dependence on AI emphasizes the need for its trustworthiness. Trustworthy AI refers to the development and deployment of AI systems that are reliable, ethical, and transparent, ensuring that they align with human values and societal norms. In this talk, I will outline our contributions to creating Trustworthy AI through both algorithmic innovations and practical applications. Firstly, I will introduce our pioneering effort that improves the robustness of Large Language Models (LLMs) through a novel perturbation consistency learning approach. Following that, I will present our work, Counterfactual Interpolation Augmentation (CIA), which offers a unified approach to enhance the fairness and explainability of Deep Neural Networks (DNNs). Additionally, I will discuss our development of Attentive Class Activation Tokens (AttCAT), designed to provide faithful token-level explanations for Transformers. In conclusion, I will explore prospective research avenues that include integrating Trustworthy AI principles into LLMs, partnering with cyber-security experts to advance Trustworthy Generative AI, and leveraging LLMs in diverse multi-modal scenarios.

Bio

Yao Qiang is a fifth-year Ph.D. student in the Department of Computer Science at Wayne State University, working in the Trustworthy AI lab under the supervision of Dr. Dongxiao Zhu. His research mainly focuses on Trustworthy AI, Natural Language Processing, Large Language Models, and Machine Learning Theory and Application. Yao's dedication to these areas has culminated in the publication of numerous research papers at the most competitive AI conferences, including NeurIPS, ICML, AAAI, IJCAI, MICCAI, IJCNN, etc. Besides his academic achievements, Yao has seven years of industry experience as a computer hardware designer and a three-month internship as an Applied Scientist at Amazon. His passion for research drives him to delve deeper into the frontiers of science and encourages him to transform theoretical discoveries into practical innovations that make a meaningful impact on society.