Xinli Xu 许新里 3rd-year PhD Candidate in Artificial Intelligence The Hong Kong University of Science and Technology (Guangzhou) | ENVISION Lab Email: xxlbigbrother@gmail.com

Biography

I am a 3rd-year PhD Candidate in Artificial Intelligence at The Hong Kong University of Science and Technology (Guangzhou) and a member of ENVISION Lab, advised by Ying-Cong Chen. Previously, I received the M.S. degree in Optical Engineering from Beijing Institute of Technology in 2023, advised by Prof. Jianan Li and Prof. Tingfa Xu, and the B.E. degree in Measurement and Control Technology and Instruments from Beijing Institute of Technology in 2020.

News

Education

PhD Candidate, Artificial Intelligence The Hong Kong University of Science and Technology (Guangzhou)	Aug. 2023 -- Present
M.S., Optical Engineering, Beijing Institute of Technology	Aug. 2020 -- June 2023
B.E., Measurement and Control Technology and Instruments, Beijing Institute of Technology	Aug. 2016 -- June 2020

Work Experience

HUAWEI Noah's Ark Lab | Beijing, China	Research Intern
Algorithm Engineer	Oct. 2022 – Jun. 2023
• Masked Autoencoders for Pre-training Large-scale Point Clouds
• FusionMAE: A Multi-Task Multi-Modality Pretraining Framework

NIO-AD (Autonomous Driving) | Beijing, China	Research Intern
Algorithm Engineer	Oct. 2021 – Aug. 2022
• 3D & TSR & Shape Autolabeling — models deployed online

Meitu Imaging & Vision Lab (MTLAB) | Beijing, China	Research Intern
Computer Vision Algorithm Engineer	Jul. 2021 – Sept. 2021
• Depth Estimation — model acceleration on iPhone and Android, ~10× speedup, deployed in APP

DXD (Robotics Startup) | Beijing, China	Founder & CEO
Founder	May 2018 – Jun. 2020
• Indoor navigation robot — hardware design and SLAM
• Awards: Beijing Outstanding Student Start-up Team; Zhongguancun High-tech Enterprise

Research Interests

My research interests include VLM, agentic multi-modal systems, and AI for creation.

Selected Honors

2020-2023, First-class Academic Scholarship, Beijing Institute of Technology

2020-2023, Outstanding Student, Beijing Institute of Technology

2020, MIIT Scholarship (40,000 RMB, Ministry of Industry and Information Technology, 工信部)

2020, Outstanding Graduate, Beijing Institute of Technology

Publications [Full list at Google Scholar]

* indicates equal contribution; † indicates project lead/corresponding author.

Preprints

• VideoMemory: Toward Consistent Video Generation via Memory Integration
  Jinsong Zhou*, Yihua Du*, Xinli Xu*, et al.
  arXiv preprint, 2026.
  
• PresentCoach: Dual-Agent Presentation Coaching through Exemplars and Interactive Feedback
  Sirui Chen*, Jinsong Zhou*, Xinli Xu*, Xiaoyu Yang, Litao Guo, Ying-Cong Chen
  arXiv preprint, 2025.
  

Journal Papers

• FusionRCNN: LiDAR-Camera Fusion for Two-stage 3D Object Detection
  Xinli Xu, Shaocong Dong, Lihe Ding, et al.
  Remote Sensing, 2022.
  

Conference Papers

• GaussianProperty: Integrating Physical Properties to 3D Gaussians with LMMs
  Xinli Xu*, Wenhang Ge*, Dicong Qiu*, et al.
  IEEE/CVF International Conference on Computer Vision (ICCV), 2025.
  [Project Page]
• FlexGen: Flexible Multi-View Generation from Text and Image Inputs
  Xinli Xu*, Wenhang Ge*, Jiantao Lin*, et al.
  IEEE/CVF International Conference on Computer Vision (ICCV), 2025.
  [Project Page]
• ComfyMind: Toward General-Purpose Generation via Tree-Based Planning and Reactive Feedback
  Litao Guo*, Xinli Xu*, Luozhou Wang, et al.
  The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025.
  [Project Page]
• Long-Video Audio Synthesis with Multi-Agent Collaboration
  Yehang Zhang*, Xinli Xu*, Xiaojie Xu*, Li Liu, Yingcong Chen
  The Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025.
  [Project Page]
• PreGenie: An Agentic Framework for High-quality Visual Presentation Generation
  Xiaojie Xu, Xinli Xu, Sirui Chen, et al.
  The Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025.
  
• FH-Net: A Fast Hierarchical Network for Scene Flow Estimation on Real-world Point Clouds
  Lihe Ding*, Shaocong Dong*, Tingfa Xu, Xinli Xu, et al.
  European Conference on Computer Vision (ECCV), 2022. (Oral, Acceptance Rate: 2.7%)
  [Code]
• MsSVT: Mixed-scale Sparse Voxel Transformer for 3D Object Detection on Point Clouds
  Shaocong Dong*, Lihe Ding*, Haiyang Wang, Tingfa Xu, Xinli Xu, et al.
  Advances in Neural Information Processing Systems (NeurIPS), 2022.
  [Code]

Invited Talks

• ComfyMind invited talk by Tencent Hunyuan, July 2025.

Teaching Assistant

2025 Fall	UCMP6050	Project Design Thinking
2024 Fall	AIAA 2025	Introduction to Artificial Intelligence

© Xinli Xu | Last updated: April 2026