Defang Chen

Defang Chen is a Researcher and Postdoctoral Associate in the Department of Computer Science and Engineering at the University at Buffalo, working under the supervision of Siwei Lyu. He is affiliated with the School of Engineering and Applied Sciences and is based at 301A Davis Hall in Buffalo, NY. His research focuses on advanced machine learning techniques, including knowledge distillation, diffusion models, graph neural networks, and domain generalization.

Defang's work emphasizes improving model efficiency through distillation methods, exploring generative models like diffusion processes, and enhancing cross-domain adaptability. His contributions span both theoretical advancements and practical applications, such as accelerating diffusion sampling and optimizing neural network architectures.

His publications between 2023–2025 highlight trends in model compression, generative AI, and graph-based learning. Notable themes include refining knowledge transfer between models, improving adversarial robustness, and developing scalable sampling techniques. Defang’s research has implications for computer vision, natural language processing, and semantic segmentation.

No academic awards or grants are explicitly listed in the provided information. While no formal advisees are noted, his role as a postdoctoral researcher suggests collaborative involvement in academic projects and teams. Defang’s work is accessible via his Google Scholar profile.