Junqi You

I am a final year undergraduate at Shanghai Jiao Tong University, majoring in Artificial Intelligence with GPA ranking top 5%. I am also a member of Guozhi honored class (held by Prof. Xiaoou Tang). Currently, I'm a research intern at RethinkLab at SJTU, advised by Prof. Junchi Yan. I also work closely with Prof. Ming-Hsuan Yang at UC Merced. Previously, I was a research intern at BCMI, advised by Prof. Li Niu.

My research interests lie in computer vision, especially at the intersection of 3D vision and generation. I also have experience in the field of autonomous driving.

[Email] / [CV] / [Github] / [Scholar]

This is my cat. Her name is Grey. She is cute.

Publications

(* denotes equal contribution)

	Efficient-LVSM: Faster, Cheaper, and Better Large View Synthesis Model via Decoupled Co-Refinement Attention Xiaosong Jia, Yihang Sun, Junqi You, Songbur Wong, Zichen Zou, Junchi Yan, Zuxuan Wu, Yu-Gang Jiang Arxiv Coming Soon* Faster and better latent feed-forward NVS model by decoupling computation for heterogeneous tokens.
	InstaInpaint: Instant 3D-Scene Inpainting with Masked Large Reconstruction Model Junqi You, Chieh Hubert Lin, Weijie Lyu, Zhengbo Zhang, Ming-Hsuan Yang NeurIPS 2025 [paper] [project page] [code] Feed-forward 3D inpainting model trained with masked self-supervision.
	Bench2Drive-R: Turning Real World Data into Reactive Closed-Loop Autonomous Driving Benchmark by Generative Model Junqi You, Xiaosong Jia, Zhiyuan Zhang, Yutao Zhu, Junchi Yan Arxiv 2025 [paper] Closed-loop simulation for E2E autonomous driving models empowered by generative model.
	DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Driving Xiaosong Jia, Junqi You, Zhiyuan Zhang, Junchi Yan ICLR* 2025 [paper] [code] Parallel attention architecture that unifies perception, mapping and planning with minimal task correlation inductive bias.
	Shadow Generation for Composite Image Using Diffusion Model Qingyang Liu, Junqi You, Jianting Wang, Xinhao Tao, Bo Zhang, Li Niu CVPR 2024 [paper] [code] Global geometry and density aware shadow generation for composite foreground object.

Publications

Recent Posts

Deep Generative Model: VAE and Diffusion