About
I am a senior student at the School of Electronic Engineering and Computer Science (EECS), Peking University. I am honored to be advised by Prof. Bin Cui and Prof. Xupeng Miao.
My research interest is broadly in Machine Learning Systems, with a particular focus on world model system acceleration. My research objective is to build efficient and scalable systems that unlock the next generation of large-scale AI workloads.
Be yourself, and be the best version of yourself.
You can also find me on GitHub.
News
Featured in People's Daily as the first listed undergraduate representative in the AY 2024-2025 National Scholarship recipients roster.
One paper was accepted by ICML 2026.
Selected as the only candidate from the School of EECS to run for Peking University Person of the Year.
Received the National Scholarship for the third time.
Received the SenseTime Scholarship.
One paper was accepted by CVPR 2025.
Awarded Beijing Municipal Triple-A Student as the only recipient in the college this year.
Received the North Star Scholarship from Palm Avenue Institution.
Received National Natural Science Foundation support for research on embodied intelligence in multimodal large models.
Received the National Scholarship for the second time.
One paper was accepted by NeurIPS 2024.
Selected for the Peking University School of Computer Science Elite Program.
Received the National Scholarship for the first time.
Selected Publications
View All →PolySplat: Workload-Regime-Aware Rasterization for 3D Gaussian Splatting
Longzan Luo, Bin Cui, Xupeng Miao
Under Review at the 40th Conference on Neural Information Processing Systems (NeurIPS 2026)
PolySplat is a workload-regime-aware 3D Gaussian Splatting rasterizer combining adaptive tile-key emission, asynchronous shared-memory staging, and persistent scheduling. It delivers geometric-mean speedups of 1.26x-6.48x over state-of-the-art rasterizers at lossless visual quality across extreme resolutions and primitive counts.
DARTS: Distribution-Aware Active Rollout Trajectory Shaping for Accelerating LLM Reinforcement Learning
Yujie Wang, Siwei Chen, Longzan Luo, Xinyi Liu, Xupeng Miao, Fangcheng Fu, Bin Cui†
Proceedings of the 43rd International Conference on Machine Learning (ICML)
DARTS introduces active distribution shaping for LLM reinforcement learning rollouts via distribution-aware trajectory sampling and adaptive redundancy allocation, delivering up to 1.77x end-to-end acceleration over state-of-the-art RL systems without compromising model quality.
Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation
Yueru Jia, Jiaming Liu†, Sixiang Chen, Chenyang Gu, Zhilue Wang, Longzan Luo, Lily Lee, Pengwei Wang, Renrui Zhang†, Zhongyuan Wang, Shanghang Zhang†
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
We propose the Lift3D framework, which progressively enhances 2D foundation models with implicit and explicit 3D robotic representations to construct a robust 3D manipulation policy.
GarmentLab: A Unified Simulation and Benchmark for Garment Manipulation
Haoran Lu, Ruihai Wu, Yitong Li, Sijie Li, Ziyu Zhu, Chuanruo Ning, Yan Shen, Longzan Luo, Yuanpei Chen, Hao Dong
Advances in Neural Information Processing Systems (NeurIPS)
GarmentLab provides realistic simulation for diverse garments with different physical properties, benchmarking novel garment manipulation tasks in both simulation and the real world.