I am a first-year Ph.D student of R&L Group at Nanjing University (NJU), under the supervision of Prof. Qi Fan โ homepage. I obtained my M.S. in Computer Science at the University of Chinese Academy of Sciences (UCAS) in 2024 and B.S at Shanghai Jiao Tong University (SJTU) in 2021. I was also fortunate to be an internship at 01AI 
, Huawei 
, TeleAI 
, Kuaishou-Kling 
, Tecent-Hunyuan.
My research interests lie in the intersection of Computer Vision and Machine Learning. From 2021, I started to do some research on Neural architecture search and image caption. Now, I focus on designing novel applications for image/video generation, World model, 3D autoregressive-generation and other downstream AIGC tasks. Welcome to the Zhihu homepage for academic discussions in the field of image/video generation.
๐ Educations
- 2024.10 - up-to-now, Phd, School of Intelligence Science and Technology, Nanjing University.
- 2021.09 - 2024.06, M.S. degree, School of Computer Science and Technology, University of Chinese Academy of Sciences.
- 2017.09 - 2021.06, B.S. degree, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University.
๐ Publications/Preprints - main contribution
- ArXiv 2026 Pathwise Test-Time Correction for Autoregressive Long Video Generation. Xunzhi Xiang, Zixuan Duan, Guiyu Zhang, Haiyu Zhang, Zhe Gao, Junta Wu, Shaofeng Zhang, Tengfei Wang, Qi Fan, Chunchao Guo. [paper] [project] [code]
- Technical Report 2025 TeleWorld: Towards Dynamic Multimodal Synthesis with a 4D World Mode. TeleWorld Team. [paper]
- ArXiv 2025 Denoising Vision Transformer Autoencoder with Spectral Regularization. Xunzhi Xiang, Xingye Tian, Guiyu Zhang, Yabo Chen, Shaofeng Zhang, Xuebo Wang, Xin Tao, Qi Fan. [paper]
- ArXiv 2025 Macro-from-Micro Planning for High-Quality and Parallelized Autoregressive Long Video Generation. Xunzhi Xiang, Yabo Chen, Guiyu Zhang, Zhongyu Wang, Zhe Gao, Quanming Xiang, Gonghu Shang, Junqi Liu, Haibin Huang, Yang Gao, Chi Zhang, Qi Fan, et al. [paper] [project] [code]
- ArXiv 2025 Make It Efficient: Dynamic Sparse Attention for Autoregressive Image Generation. Xunzhi Xiang, Qi Fan. [paper]
- AAAI 2025 ReMask-Animate: Refined Character Image Animation Using Mask-Guided Adapters. Xunzhi Xiang, Haiwei Xue, Zonghong Dai, Di Wang, Minglei Li, Ye Yue, Fei Ma, Weijiang Yu, Heng Chang, Fei Richard Yu. [paper]
- AISTATS 2025 A Neural Architecture Predictor based on GNN-Enhanced Transformer. Xunzhi Xiang, Kun Jing, Jungang Xu. [paper]
๐ฅ Publications/Preprints - participating contribution
- CVPR 2026 HERA: Efficient Test-Time Adaptation for Cross-Domain Few-Shot Segmentation with Vision Foundation Models. Junyuan Ma, Xunzhi Xiang, Wenbin Li, Yang Gao, Qi Fan. [paper]
- CVPR 2026 SymphoMotion: Joint Control of Camera Motion and Object Dynamics for Coherent Video Generation. Guiyu Zhang, Yabo Chen, Xunzhi Xiang, Junchao Huang, Zhongyu Wang, Li Jiang. [paper]
- ICLR 2026 Retain and Adapt: Auto-Balanced Model Editing for Open-Vocabulary Object Detection under Domain Shifts. Zixuan Duan, Fengyuan Lu, Xunzhi Xiang, Wenbin Li, Yang Gao, Qi Fan. [paper]
- ICLR 2026 QPrompt-R1: Real-Time Reasoning for Domain-Generalized Semantic Segmentation via Group-Relative Query Alignment. Fengyuan Lu, Zixuan Duan, Xunzhi Xiang, Zhicheng Zhang, Wenbin Li, Yang Gao, Qi Fan. [paper]
- NeurIPS 2025 DONโT NEED RETRAINING: A Mixture of DETR and Vision Foundation Models for Cross-Domain Few-Shot Object Detection. Chang-han Liu, Xunzhi Xiang, Zixuan Duan, Wenbin Li, Yang Gao, Qi Fan. [paper]
- SIGGRAPH 2025 Proteus-ID: ID-Consistent and Motion-Enhanced Video Customization. Guiyu Zhang, Chen Shi, Zijian Jiang, Xunzhi Xiang, Jingjing Qian, Shaoshuai Shi, Li Jiang. [paper] [project]
- TPAMI 2025 Human Motion Video Generation: A Survey. Haiwei Xue, Xiangyang Luo, Zhanghao Hu, Xin Zhang, Xunzhi Xiang, Yuqing Dai, Jianzhuang Liu, Zhensong Zhang, Minglei Li, Jian Yang, Fei Ma, Changpeng Yang, Zonghong Dai, Fei Richard Yu. [paper]
- ArXiv 2025 SmartSAM: Segment Ambiguous Objects like Smart Annotators. Zhe Gao, Shiyu Shen, Xunzhi Xiang, Wenbin Li, Yang Gao, Qi Fan. [paper]
๐ Honors and Awards
- 2021.10 CCF-BDCI award.