Haoqin Hong completed his undergraduate studies in 2025 at the Hanhong Honors College, Yuan-Longping Program at Southwest University, earning a B.Eng in Computer Science and Technology under the guidance of Prof. Xiaofang Hu. Starting in September 2025, he will pursue an M.Phil degree in Artificial Intelligence at the School of Artificial Intelligence and Data Science, USTC, supervised by Prof. Jingrun Chen (NSFC Distinguished Young Scholar), and he will also collaborate with Associate Researcher Congcong Zhu on Computer Vision. He did research internships at companies or institute such as iFLYTEK Research.

Haoqin Hong's research interests focus on physics-inspired robotic perception and world models. Currently, he is a Kaggle Expert, served as a program committee member for the AAAI international conference, and as a reviewer for international conferences and journals such as IJCAI and IJCNN. Familiar with the fundamental architectures and cutting-edge research developments of algorithms, including Transformer, Diffusion, VLA, NeRF, 3DGS, Deep Reinforcement Learning, and Feed-Forward Models.

• CV(English)   CV(Chinese)

Github Haoqin Hong's Github

Email     HaoqinHong@outlook.com  haoqinhong@mail.ustc.edu.cn   honghaoqin@email.swu.edu.cn

Google Scholar Haoqin Hong's Google Scholar

Address Tianchen Road, High-tech West District, Chengdu, Sichuan Province, China, 611731

Recent Work

News

• 2025.9   🎉 My coauther paper was accepted by Neurocomputing.
• 2025.4   🎉 I was selected as an Outstanding Graduate in Southwest University.
• 2024.9   🎉 I will start my M.Phil in Intelligent Science and Technology at USTC.
• 2024.7   🎉 I present at the World Congress on Computational Intelligence held in Yokohama, Japan.
• 2024.3   🎉 My first-author paper was accepted by IEEE IJCNN 2024.
• 2023.6   🎉 I have been admitted to the China Academy of Sciences Innovation Training Program.

Publications

Articles (^* Equal contribution, ^† Corresponding author, ^‡ Project leader)

1. Physics-Informed Deformable Gaussian Splatting: Towards Unified Constitutive Laws for Time-Evolving Material Field. [Project]
   Haoqin Hong^*‡, Din Fan^*, Fubing Dou, Zhili Zhou, Haoran Sun, Congcong Zhu^† , Jingrun Chen.
   Under Review.  
2. CCSPNet-Joint: Efficient Joint Training Method for Traffic Sign Detection Under Extreme Conditions. [Link]
   Haoqin Hong, Yue Zhou^† , Xiangyu Shu, Xiaofang Hu.
   Proceedings of the 2024 IEEE International Joint Conference on Neural Networks, Yokohama, Japan.   IJCNN & WCCI 2024 Oral CCF-C Conference
3. Attaining Hardware-Efficient Inference in Memristor-Based Transformer Accelerator via Network Redesign.
   Junze Xu, Haoqin Hong, Yue Zhou^† , Xiaofang Hu, Shukai Duan.
   Neurocomputing.   CAS SCI Q2, IF 5.5 CCF-C Journal

Chinese Invention Patent

1. Speech Recognition System and Method Based on Lightweight Transformer Network.
2. Text Sentiment Detection System and Method Based on Memristor.

Selected Awards

Honors and Awards

1. Outstanding Individual in Innovation Capability in Chongqing.
2. Outstanding Individual in Enhancing Innovation Ability in Chongqing.
3. Outstanding Graduates in Southwest University.
4. Outstanding Undergraduate Thesis of Southwest University.
5. Southwest University Hanhong Foundation Student Growth and Success Scholarship.
6. Southwest University First Class Scholarship.

Competition Awards

1. 🥉 Bronze Medal in Kaggle Research Competition: NeurIPS-OpenPolymerPrediction 2025.
2. 🥉 Bronze Medal in Kaggle Research Competition: CVPR 2025 Workshop - Image Matching Challenge.
3. 🥉 Bronze Medal in Kaggle Research Competition: Google - Fast or Slow? Predict AI Model Runtime.
4. 🏅 National First Prize in the DJI RoboMaster of the National University Student Robot Competition.
5. 🏅 National First Prize in the China Robotics and Artificial Intelligence Competition.
6. 🥈 National Second Prize in the China/Contemporary Undergraduate Mathematical Contest in Modeling.
7. 🏅 Provincial First Prize in the National Undergraduate Electronics Design Contest.

Projects

• Image Processing Based on Multimodal Astronomical Large Models. [Certificate]
  Chinese Academy of Sciences Innovation Program Leader, Funded by the National Astronomical Observatories of the Chinese Academy of Sciences, and Advised by Prof. Ali Luo.
• Attention Networks based on Memristors in Natural Language Processing. [Certificate]
  Provincial Innovation Training Program Leader.
• Real-Time Robot Object Detection and SLAM 3D Map Building and Navigation.
  Leader of Algorithm Group, Robotics Innovation Lab, SWU.

Representative Work Showcase

IJCNN & WCCI Oral

CCSPNet-Joint: Efficient Joint Training Method for Traffic Sign Detection Under Extreme Conditions.
IEEE IJCNN 2024 Oral & WCCI 2024
Project:    Github Code       Kaggle Dataset
Authors:   Haoqin Hong @SWU, Yue Zhou* @SWU, Xiangyu Shu @SWU, Xiaofang Hu @SWU.

DJI RoboMaster

Real-Time Robot Object Detection and SLAM 3D Map Building and Navigation.
Head of Vision Group, Robotics Innovation Lab, SWU
Project:    Github Object Detection     Github SLAM Navigation
Current status of team members:    Chengyu Fang @Tsinghua, Haoqin Hong @USTC, Bohan Ma @ZJU, Xiangyu Shu @SCUT, Yangce Xiong @UWA, Jiesen Zhao @BHU, Xinke Zhang @XidianU et al.
Keywords:    Robotics Vision, SLAM Navigation.