News-2023

Invited to serve as the Area Chair for the conference ICLR’24. (Dec. 2023)
Our paper “The Emergence of Reproducibility and Consistency in Diffusion Models”, has been awarded as the Best Paper Award at NeurIPS Diffusion Model Workshop 2023. (Nov. 2023)
Elected to serve on the IEEE Computational Imaging Technical Committee (CI TC) for a 3-year term starting January 2024. (Nov. 2023)
Invited to give a talk at UMich Department of Learning Health Sciences (DLHS) Research Seminar about AI for Medical Imaging. (Nov. 2023)
Received a grant of 2023 MICDE Catalyst Grant from Michigan Institute for Computational Discovery and Engineering (MICDE) in collaboration with Prof. Qing Qu and Prof. Jeffrey A. Fessler, for supporting our research on Efficient Diffusion Models for Scientific Machine Learning. (Oct. 2023)
Invited to serve as the Associate Editor for the journal BJR|Artificial Intelligence. (Oct. 2023)
Release our new paper on arXiv: “The Emergence of Reproducibility and Consistency in Diffusion Models”. Short version is selected as an oral (top 2%) at NeurIPS 2023 Workshop on Diffusion Models. (Oct. 2023)
Proposal “Develop Efficient Diffusion Models for Solving General Inverse Problems” has been accepted as a Discover project in the NSF ACCESS program.
Release our new paper on arXiv: “Poisson-Gaussian Holographic Phase Retrieval with Score-based Image Prior”. Short version is accepted at NeurIPS 2023 Deep Learning and Inverse Problems Workshop. (Sep. 2023)
Invited to give talks at ICCV 2023 Computer Vision for Automated Medical Diagnosis (CVAMD) Workshop, UMich MIDAS mini-symposium on AI for Medical Imaging, UMich ECE Communications and Signal Processing Seminar about our recent work using latent diffusion models for solving general inverse problems. Please check the video recording. (Sep. 2023)
Organize research symposium at U-M MIDAS: “Generative AI: Diffusion Models for Scientific Machine Learning”. (Sep. 2023)
Release our new paper on arXiv: Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistency. (Jul. 2023)
Received a grant of Propelling Original Data Science (PODS) 2023 award from Michigan Institute for Data Science (MIDAS) in collaboration with Prof. Lise Wei (UMich Medicine), for supporting our collaborative research on personalized progression risk prediction for cancer patients from longitudinal multi-modality data. (Jun. 2023)
Proposal “Develop Computationally Efficient Algorithms Solving Inverse Problems with Latent Diffusion Models” has been accepted as an Explore project in the NSF ACCESS program.
Organization committee for a new conference: Conference on Parsimony and Learning (CPAL), which will be held in Jan. 2024.