Duke Computer Science Alumni Win Best Paper Award at MICRO 2025

Duke Computer Science alumni Entropy Xu and Yongji Wu, along with professors Danyang Zhuo, Matthew Lentz, Lisa Wu Wills and collaborators at Carnegie Mellon University, won the Best Paper Award at the MICRO 2025 conference on October 20th in Seoul, South Korea.  

The MICRO IEEE/ACM International Symposium on Microarchitecture^® is the premier forum for presenting, discussing, and debating innovative microarchitecture ideas and techniques for advanced computing and communication systems.  

The award-winning paper, “LLM.265: Video Codecs are Secretly Tensor Codecs,” was one of the two best papers of the 597 that were submitted this year.  

Entropy Xu came to Duke in 2020, where he primarily focused on the intersection between computer architecture and machine learning under the guidance of Lisa Wills. During his time on campus he authored numerous publications, winning the ISPASS 2023 Best Paper Award. He completed his PhD. degree in Computer Science in 2024.   

Yongji Wu also began graduate studies at Duke in 2020 after earning an undergraduate degree at the University of Science and Technology of China. With a focus on Machine Learning and Datacenter Networks, Wu worked with Professors Zhuo and Lentz, focusing on building systems for optimized performance and managing learning networks. Currently, Yongji Wu is conducting postdoctoral research at the University of California, Berkeley, further exploring his passions for research in machine learning.  

View Entropy’s presentation of the team’s work at MICRO 2025.