Frontier of Artificial Network
A Series of Invited Talks @ FAN Group, CityU

Biography: Zijun Cui is an Assistant Professor at Michigan State University. Prior to that, she was a postdoctoral research fellow at the University of Southern California. She received the Ph.D. degree from the Department of ECSE at Rensselaer Polytechnic Institute in 2022. Dr. Cui has worked on deep learning, computer vision, and probabilistic graphical models. Her research interests lie in leveraging domain knowledge to advance deep learning, include physics-informed deep learning and neural-symbolic integration. Her research applications include computer vision, healthcare, and natural language processing. Her work appears in conferences such as CVPR, NeurIPS, AAAI, IJCAI, UAI, AISTATS, MICCAI and journals such as IEEE Transaction on Neural Networks and Learning Systems and npj Digital Medicine. From 2019 to 2022, she was awarded the Rensselaer-IBM Artificial Intelligence Research Collaboration Scholarship. Learn more about Dr. Cui at https://zijunjkl.github.io/.

Abstract: Current artificial intelligence (AI) has made substantial progress across various fields, such as computer vision and natural language processing. However, limitations of current AI persist in its dependency on data quantity and quality, lack of transparency, and absence of general intelligence. These issues become particularly apparent when considering the adaptability and comprehension of humans. My research in AI + Knowledge aims to leverage valuable domain-specific human expertise for advanced AI, seeking to enhance the data efficiency, generalizability, and interpretability of existing deep learning models. In this talk, I will begin by introducing my research framework on AI + Knowledge, providing insight into its three major tasks, applications, and my contributions. I will then delve into several use cases to introduce novel AI + Knowledge techniques and demonstrate their impact in domains like computer vision, healthcare, and beyond. Specifically, I will highlight the significance of leveraging biomechanics to improve both the accuracy and interpretability of human motion understanding covering face, body, and hand in both 2D and 3D. Concluding the talk, I will outline future research directions, which include advancing synergy between domain knowledge and AI, fostering the continual growth of both knowledge and AI, and adapting AI + Knowledge techniques to address the real-world challenges.

Biography: Yang Cao is an Associate Professor at the Department of Computer Science, Institute of Science Tokyo (Science Tokyo, formerly Tokyo Tech), and directing the Trustworthy Data Science and AI (TDSAI) Lab. He is passionate about studying and teaching on algorithmic trustworthiness in data science and AI. Two of his papers on data privacy were selected as best paper finalists in top-tier conferences IEEE ICDE 2017 and ICME 2020. He was a recipient of the IEEE Computer Society Japan Chapter Young Author Award 2019, Database Society of Japan Kambayashi Young Researcher Award 2021. His research projects were/are supported by JSPS, JST, MSRA, KDDI, LINE, WeBank, etc.

Abstract: Large language models are often fine-tuned on small and sensitive datasets, where individual training examples can strongly influence model behavior and lead to memorization and privacy leakage. This talk focuses on differential privacy in LLM fine-tuning and explains how DP acts as a learning constraint that limits the influence of individual samples, thereby mitigating such risks. We introduce the core intuition behind DP without assuming prior background, review practical DP fine-tuning mechanisms, and discuss how privacy and utility should be evaluated together. We also clarify what DP fine-tuning can and cannot protect, and outline open challenges in privacy-aware LLM adaptation.

Biography: Jiancheng Yang is a Principal Investigator at the ELLIS Institute Finland and Assistant Professor at Aalto University. He received his Bachelor's and PhD degrees from Shanghai Jiao Tong University, was a visiting fellow at Harvard, and completed postdoctoral training at EPFL. His research advances AI for health, with a focus on spatial intelligence, generative AI, and multimodal deep learning. He has authored 60+ publications and is recognized for developing MedMNIST. He is a Forbes 30 Under 30 honoree and a recipient of the WAIC Yunfan Award.

Abstract: The next step for medical AI is scaling. However, scaling medical AI faces intrinsic challenges: medical data are often limited, complex, and highly heterogeneous, while many existing AI techniques are not yet designed to fully exploit such data. This talk presents contributions unified by a common goal: building scalable medical AI through maximal reuse. First, I introduce LeFusion, a constrained generative framework for medical data synthesis (ICLR 2025 Spotlight), which provides guarantees on synthetic data quality and allows downstream models to better exploit underutilized information in limited real datasets. Second, I describe DiffAtlas, a new perspective on multimodal medical data (MICCAI 2025 Spotlight), where traditional input–output medical image segmentation is reformulated as an unconditional multimodal diffusion model guided only at inference time. Finally, I introduce PlaneCycle, a training-free, adapter-free operator that lifts arbitrary 2D foundation models into 3D by cyclically distributing spatial aggregation across orthogonal planes, adding no parameters while preserving pretrained inductive biases. Together, these works suggest a broader principle: medical AI can scale without scaling cost.