Scholar
Haozhe Liu
Google Scholar ID: QX51P54AAAAJ
KAUST
Citations & Impact
Citations
949
H-index
18
i10-index
21
Publications
20
Co-authors
29
list available
Publications
Resume
Academic Achievements
- - Published over 10 papers in top-tier venues, with more than 700 citations on Google Scholar
- - Notable projects: BoxDiff, NLSOM, TGATE, MarDini
- - Co-developed the pretraining of a 7B commercial text-to-image model at Meta, scaling to O(1B) training samples
- - Papers accepted by TMLR, CVPR, ICCV, ECCV, MICCAI, AAAI
- - NLSOM recognized as the best paper at NeurIPS 2023 workshop
Research Experience
- - Research Scientist Intern at Meta AI (London), focusing on efficient video generation (Summer 2024)
- - Intern at Tencent (Shenzhen), focusing on GAN-based image synthesis and medical imaging
- - Research Scientist Intern at Meta (MPK), focusing on GenAI-related topics (Summer 2025)
Education
- - Degree: Ph.D. Candidate
- - University: KAUST
- - Advisor: Prof. Juergen Schmidhuber
- - Time: August 2022 - Present
- - Field: Not specified
Background
- - Research Interests: Multimodal generative models, particularly video and image generation
- - Long-term goal: Developing a Physical AI Model (world model)
- - Brief: Currently pursuing a Ph.D. at KAUST, supervised by Prof. Juergen Schmidhuber. Previously interned at Meta AI and Tencent.
Miscellany
- - Open to future collaborations, whether it's co-founding a venture or pursuing full-time opportunities in industry or academia
- - Personal interests not specified
