PhD Student, Imperial College London

Contact: oliveryanzuolu AT gmail DOT com

<- Click my avatar to find me :)

Google Scholar     GitHub     LinkedIn     CV

👋 Hi, this is Yanzuo Lu, a PhD student at Imperial College London supervised by Jiankang Deng. My current research focuses on building Real-Time and Long-Video World Models, involving three main areas: (1) real-time generation, with a focus on achieving inference speeds where the processing time for each video chunk is less than its duration; (2) long-video generation, with the key challenge being the mitigation of error accumulation between chunks to maintain long-term consistency; (3) interactive world model, with the objective of developing methods for effective control signal injection and responsive, timely feedback.

My Chinese name is 卢彦作, and you may also want to call my English name as Oliver Lu. I love watching anime and playing video games in my spare time. My favorite anime include Demon Slayer: Kimetsu no Yaiba (鬼滅の刃), Attack on Titan (進撃の巨人), Frieren: Beyond Journey's End (葬送のフリーレン), Eighty Six (86―エイティシックス―) and Violet Evergarden (ヴァイオレット・エヴァーガーデン). Recently I've also been trying to get into landscape photography.

Feel free to explore my works below and reach out via email for any discussion.

PhD Student, Imperial College London

Oct 2025 - Present, London, United Kingdom

In Department of Computing, Supervised by Jiankang Deng;

Research Topic: real-time and long-video world models

Research Intern, ByteDance Seed

Dec 2023 - Sep 2025 (1 yr 10 mo), Shenzhen, China

Mentored by Yuxi Ren & Jie Wu and led by Xuefeng Xiao;

Research Topic: accelerating diffusion model to reduce sampling steps via progressive/consistency/rectified/score/adversarial distillation and RLHF for efficient image and video synthesis;

Industry Deployment: Douyin/TikTok (short-form content), Capcut (video editor), Dreamina (image & video generator), Doubao (chatbot)

Master of Engineering (MEng), Sun Yat-Sen University

Sep 2022 - Jun 2025, Guangzhou, China

In School of Computer Science and Engineering;

Supervised by Andy J Ma, Xiaohua Xie, and Jianhuang Lai;

Research Topic: customized diffusion models, domain adaptation and person re-identification

Bachelor of Engineering (BEng), Sun Yat-Sen University

Sep 2018 - Jun 2022, Guangzhou, China

In School of Computer Science and Engineering;

Relevant Coursework: Probability and Statistics, Machine Learning and Data Mining, Principles of Artificial Neural Networks, Optimization Theory, Artificial Intelligence, Computer Vision, Computer Graphics, etc.

Adversarial Distribution Matching for Diffusion Distillation Towards Efficient Image and Video Synthesis
Yanzuo Lu, Yuxi Ren, Xin Xia, Shanchuan Lin, Xing Wang, Xuefeng Xiao, Andy J Ma, Xiaohua Xie and Jianhuang Lai
International Conference on Computer Vision (ICCV), 2025 (Highlight)
[arXiv]     [Publication]

Coarse-to-Fine Latent Diffusion for Pose-Guided Person Image Synthesis
Yanzuo Lu, Manlin Zhang, Andy J Ma, Xiaohua Xie and Jianhuang Lai
Conference on Computer Vision and Pattern Recognition (CVPR), 2024 (Highlight)
[arXiv]     [Publication]     [Code (200+ stars on GitHub)]     [Talk]

Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image Synthesis
Yuxi Ren, Xin Xia, Yanzuo Lu, Jiacheng Zhang, Jie Wu, Pan Xie, Xing Wang and Xuefeng Xiao
Conference on Neural Information Processing Systems (NeurIPS), 2024
[arXiv]     [Publication]     [Project Page]     [HuggingFace (Over 4M downloads)]     [PR]

ByteEdit: Boost, Comply and Accelerate Generative Image Editing
Yuxi Ren, Jie Wu, Yanzuo Lu, Huafeng Kuang, Xin Xia, Xionghui Wang, Qianqian Wang, Yixing Zhu, Pan Xie, Shiyin Wang, Xuefeng Xiao, Yitong Wang, Min Zheng and Lean Fu
European Conference on Computer Vision (ECCV), 2024
[arXiv]     [Publication]     [Project Page]

Seedream 4.0: Toward Next-generation Multimodal Image Generation
ByteDance Seed (Core Contributor)
arXiv preprint arXiv:2509.20427, 2025
[arXiv]     [Project Page]

Hyper-Bagel: A Unified Acceleration Framework for Multimodal Understanding and Generation
Yanzuo Lu, Xin Xia, Manlin Zhang, Huafeng Kuang, Jianbin Zheng, Yuxi Ren, Xuefeng Xiao
arXiv preprint arXiv:2509.18824, 2025
[arXiv]     [Project Page]

Seedream 3.0 Technical Report
ByteDance Seed (Contributor)
arXiv preprint arXiv:2504.11346, 2025
[arXiv]     [Project Page]