Xuan Ju

Ph.D. Candidate, The Chinese University of Hong Kong
Multimodal, Microsoft AI

E-mail CV Google Scholar GitHub LinkedIn

Biography

Hi, I'm Xuan Ju (鞠璇). I am a final-year Ph.D. student at The Chinese University of Hong Kong and am also working on image generation at Microsoft AI. My research interests broadly span multimodal learning, with a particular focus on image and video generation. Please do not hesitate to reach out to me via email at juxuan.27@gmail.com!

News

• [08/2026] We release MAI-Image-2.6, ranked No. 2 on the Arena text-to-image leaderboard.
• [06/2026] We release MAI-Image-2.5, ranked No. 3 for text-to-image and No. 2 for image-editing.
• [03/2026] We release MAI-Image-2, pushing MAI into the top 3 text-to-image labs.
• [01/2026] EditVerse is accepted to ICLR 2026 as an Oral Presentation!
• [10/2025] Join Microsoft AI as a Member of Technical Staff, building the MAI-Image series from the ground up.
• [09/2025] We release a unified image/video editing/generation framework, EditVerse.
• [06/2025] FullDiT is accepted to ICCV 2025, see you in Hawai'i!
• [05/2025] We are hosting the CVPR 2025 Workshop on Efficient and On-Device Generation (EDGE).
• [04/2025] Start an internship in Adobe Research with Soo Ye Kim and Zhe Lin.
• [04/2025] VideoPainter and Cobra are accepted to SIGGRAPH 2025.

Product Releases

MAI-Image Series Microsoft AI · Oct. 2025 - Present

Microsoft AI's in-house image generation model family, built from the ground up, powering image generation in Copilot and Bing Image Creator.

Mar. 2026

MAI-Image-2

Arena Top-3 Lab

Jun. 2026

MAI-Image-2.5

No. 3 Text-to-Image
No. 2 Image Edit

Aug. 2026

MAI-Image-2.6

No. 2 Text-to-Image

Selected Publications

* equal contribution

EditVerse: Unifying Image and Video Editing and Generation with In-Context Learning

Xuan Ju, Tianyu Wang, Yuqian Zhou, He Zhang, Qing Liu, Nanxuan Zhao, Zhifei Zhang, Yijun Li, Yuanhao Cai, Shaoteng Liu, Daniil Pakhomov, Zhe Lin, Soo Ye Kim, Qiang Xu.

International Conference on Learning Representations (ICLR), 2026 (Oral)

Project arXiv Code Benchmark Slides Full Comparison

FullDiT: Multi-Task Video Generative Foundation Model with Full Attention

Xuan Ju, Weicai Ye, Quande Liu, Qiulin Wang, Xintao Wang, Pengfei Wan, Di Zhang, Kun Gai, Qiang Xu.

International Conference on Computer Vision (ICCV), 2025

Project arXiv Video Data

MiraData: A Large-Scale Video Dataset with Long Durations and Structured Captions

Xuan Ju*, Yiming Gao*, Zhaoyang Zhang*, Ziyang Yuan, Xintao Wang, Ailing Zeng, Yu Xiong, Qiang Xu, Ying Shan.

Conference on Neural Information Processing Systems (NeurIPS), 2024

Project arXiv Video Code Data

BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion

Xuan Ju, Xian Liu, Xintao Wang, Yuxuan Bian, Ying Shan, Qiang Xu.

European Conference on Computer Vision (ECCV), 2024

Project arXiv Video Code Data

PnP Inversion: Boosting Diffusion-based Editing with 3 Lines of Code

Xuan Ju, Ailing Zeng, Yuxuan Bian, Shaoteng Liu, Qiang Xu.

International Conference on Learning Representations (ICLR), 2024

Project arXiv Video Code Data

HumanSD: A Native Skeleton-Guided Diffusion Model for Human Image Generation

Xuan Ju*, Ailing Zeng*, Chenchen Zhao*, Jianan Wang, Lei Zhang, Qiang Xu.

IEEE/CVF International Conference on Computer Vision (ICCV), 2023 (Oral)

Project arXiv Video Code Data

Human-Art: A Versatile Human-Centric Dataset Bridging Natural and Artificial Scenes

Xuan Ju*, Ailing Zeng*, Jianan Wang, Qiang Xu, Lei Zhang.

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023

Project arXiv Video Code Data

Experiences

Multimodal, Microsoft AIOct. 2025 - Present

Building the MAI-Image series from the ground up

Research Intern, Adobe ResearchMar. 2025 - Sep. 2025

Multimodal Video Generation · with Soo Ye Kim and Zhe Lin

Research Intern, Kuaishou KlingDec. 2024 - Mar. 2025

Multimodal Video Generation · with Xintao Wang

Research Intern, Meta GenAI (Llama Applied)Jul. 2024 - Dec. 2024

Llama 4 Native Image Generation · with Kevin Chih-Yao Ma

Research Intern, Tencent ARC LaboratoryOct. 2023 - Jun. 2024

Image Generation & Video Data · with Xintao Wang and Ying Shan

Research Intern, International Digital Economy Academy (IDEA) CVRJun. 2022 - Sep. 2023

Human-Centric Image Generation · with Ailing Zeng and Lei Zhang

Research Intern, X-Lab, SenseTime ResearchDec. 2021 - May. 2022

Human Pose Estimation · with Ailing Zeng and Lei Yang

Miscellaneous

• Professional Services: Conference Reviewer for ICLR, ECCV, IJCV, CVPR, ICCV, NeurIPS, AAAI, AISTATS, ACCV, WACV, etc.; Organizer of the Workshop on Efficient and On-Device Generation (EDGE) at CVPR 2025 and CVPR 2026.
• Honors & Awards: National Scholarship (2020, 2021), Outstanding Graduate of Shanghai Province (2022), Qidi Scholarship (2022), First Prize in "Challenge Cup" Contest (2022).
• Teaching: Teaching Assistant at CUHK: CSCI 1120 (Fall 2023), CSCI 1540 (Fall 2022), CSCI 1520 (Spring 2022).