|
Ph.D. Student, Texas A&M University Spatial Intelligence · 3D/4D World Modeling · Embodied Intelligence Homepage · Google Scholar · CV · Email |
I work toward systems that can perceive, decide, and act in the physical world. My work covers 3D vision-language models, semantic 3D reconstruction, and 4D world modeling, and I am increasingly focused on intelligence for embodied systems.
* denotes equal contribution.
SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning
Jian Zhang*, Shijie Zhou*, Bangya Liu*, Achuta Kadambi, Zhiwen Fan · CVPR 2026
Code · Project · Model · Data
VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction
Zhiwen Fan*, Jian Zhang*, et al. · CVPR 2026
Code · Project
Large Spatial Model: End-to-end Unposed Images to Semantic 3D
Zhiwen Fan*, Jian Zhang*, et al. · NeurIPS 2024
Code · Project
Also co-author of:
- Thinking in Dynamics: How Multimodal Large Language Models Perceive, Track, and Reason Dynamics in Physical 4D World · CVPR 2026 · Code · Project
- DynamicVerse: A Physically-Aware Multimodal Framework for 4D World Modeling · NeurIPS 2025 · Code · Project
- InstantSplat: Sparse-view Gaussian Splatting in Seconds · arXiv 2024 · Code · Project
Full list on Google Scholar.
- Ph.D., Computer Engineering · Texas A&M University · Sep 2026 – present
Advisor: Zhiwen Fan - M.S., Information & Communication Engineering · Xiamen University · Sep 2023 – Jun 2026
Advisors: Yue Huang, Xinghao Ding - B.S., Data Science and Big Data Technology · Nanchang University · Sep 2019 – Jun 2023
Advisor: Li Zhu
- Remote Research Assistant · VITA Group, University of Texas at Austin · Jan 2024 – May 2025
3D semantic reconstruction. Advisor: Atlas Wang
Open to research collaborations and internships. Email: zjrandomyeah@gmail.com
