Hongye Xu

Ph.D. Student in Imaging Science, Rochester Institute of Technology

prof_pic.jpg

Center for Imaging Science

Rochester Institute of Technology

Machine Learning · Computer Vision · Continual Learning

I am a Ph.D. student in Imaging Science at Rochester Institute of Technology, where I am advised by Prof. Bartosz Krawczyk. My research focuses on machine learning, computer vision, and continual learning.

I am interested in building adaptive and reliable visual learning systems that can learn from evolving data streams without catastrophic forgetting. Recently, I have been working on exemplar-free class-incremental learning, prototype-based continual learning, and parameter-efficient adaptation of pretrained vision models.

My broader research interests include continual learning, computer vision, efficient adaptation, vision-language learning, and trustworthy AI systems.

news

May 27, 2026 I received the ICML 2026 Gold Reviewer Award.
May 12, 2026 Our paper “Revisiting Prototype Rehearsal for Exemplar-Free Continual Learning: Manifold-Aware Boundary Sampling with Adaptive Class-Balanced Loss” was accepted to CVPR 2026 Findings.
Jan 21, 2026 Our paper “Two-Way Is Better Than One: Bidirectional Alignment with Cycle Consistency for Ex-emplar-Free Class-Incremental Learning” was accepted to ICLR 2026.

selected publications