About

I am an Applied Research Scientist at Amazon AGI Foundation, working on the Nova 2 project, focusing on multimodal and visual generation.

I received my Ph.D. from the College of Engineering at Northeastern University, advised by Prof. Yun Raymond Fu in the SMILE Lab. Before that, I spent two years in the Department of Computer Science & Engineering at the University of North Texas, working with Prof. Song Fu. I received my B.S. and M.S. degrees from Nanjing Forestry University, China.

My research interests include Model Efficiency Multimodal LLMs Generative AI Visual Generation

I did internships at Bell Labs eBay Microsoft Adobe Meta GenAI Thanks to all my mentors and collaborators for the guidance and help, always appreciate.

Experience

Amazon AGI Foundation, Applied Research Scientist 2025 – Present
Working on Nova 2 omni model, focusing on unified multimodal and visual generation. Contributing end-to-end as a main contributor across data, model design, pretraining, fine-tuning, and RL.
Meta GenAI, Research Intern 2024
Working on Llama 4 project, focusing on native visual generation exploration.
Adobe Research, Research Intern 2023
Exploring image layer-wise representation, and SVG generation with SOTA segmentation models.
Microsoft Azure, Research Intern 2023 – 2024
Exploring efficient, lightweight vision network design, focusing on-device real-time latency including GPU, CPU and iPhone.

Services

Area Chair: ECCV 2026, NeurIPS 2026 Position Track
Journal Reviewer: IEEE TPAMI, TIP, TMM, TCSVT, SPL, TIV
Conference Reviewer: CVPR, ICCV, ECCV, ICML, ICLR, NeurIPS, SIGGRAPH Asia, KDD, AAAI, IJCAI, etc.
Workshop Reviewer: ECV @ CVPR, T4V @ CVPR, AI and Social Good @ IJCAI, AMFG @ ICCV

Honors and Awards

  • 2023 Outstanding Reviewer, CVPR
  • 2022 Outstanding Reviewer, NeurIPS
  • 2022 Academic Technology Scholar, Northeastern University
  • 2020 Best Student Paper, ICME
  • 2019 Best Paper, SEC

Selected Publications [Full Google Scholar]

Nova 2
Amazon Nova 2: Multimodal Reasoning and Generation Models
Amazon Artificial General Intelligence
Technical Report 2025
Paper
Fine-T2I
Fine-T2I: An Open, Large-Scale, and Diverse Dataset for High-Quality T2I Fine-Tuning
Xu Ma, Yitian Zhang, Qihua Dong, Yun Fu
arXiv 2025
Dataset Explore Space
Token-Shuffle
Token-Shuffle: Towards High-Resolution Image Generation with Autoregressive Models
Xu Ma, Peize Sun, Haoyu Ma, Hao Tang, Chih-Yao Ma, Jialiang Wang, Kunpeng Li, Xiaoliang Dai, et al.
arXiv 2025
arXiv
Rewrite the Stars
Rewrite the Stars
Xu Ma, Xiyang Dai, Yue Bai, Yizhou Wang, Yun Fu
CVPR 2024
arXiv Code
EfficientMod
Efficient Modulation for Vision Networks
Xu Ma, Xiyang Dai, Jianwei Yang, Bin Xiao, Yinpeng Chen, Yun Fu, Lu Yuan
ICLR 2024
OpenReview Code
Image as Set of Points
Image as Set of Points
Xu Ma*, Yuqian Zhou*, Huan Wang, Can Qin, Bin Sun, Chang Liu, Yun Fu
ICLR 2023 (Oral)
Paper Code
LIVE
Towards Layer-wise Image Vectorization
Xu Ma, Yuqian Zhou, Xingqian Xu, Bin Sun, Valerii Filev, Nikita Orlov, Yun Fu, Humphrey Shi
CVPR 2022 (Oral)
arXiv Code Project Demo
PointMLP
Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework
Xu Ma, Can Qin, Haoxuan You, Haoxi Ran, Yun Fu
ICLR 2022
arXiv Code