Scholar
Xinjie Zhang
Google Scholar ID: U_gSl6wAAAAJ
Researcher, Microsoft Research Asia
Citations & Impact
Citations
418
H-index
10
i10-index
10
Publications
20
Co-authors
46
list available
Publications
Cited
0
Resume (English only)
Academic Achievements
- Released our latest unified multimodal understanding and generation foundation model Ovis-U1 (2025/06).
- Paper 'MEGA: Memory-Efficient 4D Gaussian Splatting for Dynamic Scenes' accepted to ICCV 2025 (2025/06).
- Paper 'Unified Multimodal Understanding and Generation Models: Advances, Challenges, and Opportunities' released (2025/05).
- Paper 'HarmoniCa: Harmonizing Training and Inference for Better Feature Cache in Diffusion Transformer Acceleration' accepted to ICML 2025 (2025/05).
- Paper 'Rethinking Diffusion Posterior Sampling: From Conditional Score Estimator to Maximizing a Posterior' accepted to ICLR 2025 (2025/01).
- Paper 'PvNeXt: Rethinking Network Design and Temporal Motion for Point Cloud Video Recognition' accepted to ICLR 2025 (2025/01).
- For full publication list, refer to his Google Scholar Profile.
Education
- Received B.Eng from School of Electronic Information and Engineering, South China University of Science and Technology (SCUT) in 2021; currently a PhD candidate at the Department of Electronic and Computer Engineering, Hong Kong University of Science and Technology (HKUST), supervised by Prof. Jun Zhang.
Background
- Research interests include Multimodal Understanding and Generation, Neural Data Representation & Compression, and Edge AI. Currently a PhD student at the Department of Electronic and Computer Engineering, HKUST, expected to graduate in 2025, open to both academic and industrial research positions in North America and Asia.
Miscellany
- Welcome to contact for any discussion and cooperation!
