Junhao Cai (蔡俊豪)

Junhao Cai (蔡俊豪)

M.S., Shenzhen University (2025-Present)

I am a Master's student in the College of Mechatronics and Control Engineering at Shenzhen University. My research areas include Text-to-3D generation, diffusion models, hypergraph neural networks (HGNN), and LoRA fine-tuning. I focus on developing parameter-efficient fine-tuning strategies and hypergraph-based architectures to advance multimodal learning and high-fidelity 3D content generation.

My academic excellence has been recognized with several distinctions, including being named an 🏆 Outstanding Graduate of Shenzhen University (2025, awarded to 333 out of 5,208 graduates) and receiving the 🏆 Bachelor of Honor from the School of Mechatronics and Control Engineering (2025), through sustained academic leadership, service, and interdisciplinary innovation.

Research focus Generative AI for text-to-3D, multimodal learning, and efficient adaptation.

Research Interests

  • Text-to-3D Generation
  • Diffusion Models
  • Multimodal Learning
  • Hypergraph Neural Networks
  • LoRA Fine-tuning

News

  • 2026.06🎉 One paper has been accepted to the European Conference on Computer Vision (ECCV 2026)!
  • 2026.04🎉 One paper has been accepted to the International Joint Conference on Artificial Intelligence (IJCAI 2026)!
  • 2026.02🎉 One paper has been accepted to the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026) Findings!
  • 2025.05🎉 One paper has been accepted to the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2025)!

Education

Academic Service

  • Reviewer of International Conference on Computer Vision. ICCV 2025
  • Reviewer of European Conference on Computer Vision. ECCV 2026
  • Reviewer of Association for the Advancement of Artificial Intelligence. AAAI 2026
  • Reviewer of The IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR 2026

Honors and Awards

  • 2026: 🏆 Graduate Academic Scholarship, Special Prize
  • 2025: 🏆 Outstanding Graduate, Shenzhen University
  • 2025: 🏆 Bachelor of Honor, School of Mechatronics and Control Engineering
  • 2024: 🥈 Second Prize, 15th Blue Bridge Cup Guangdong Regional Competition (Embedded Design & Development, University Group)
  • 2023: 🎖️ Outstanding Communist Youth League Member
  • 2024: 🥇 First Prize, Public Welfare Star (Academic Year 2023-2024)
  • 2024: 🏅 National Encouragement Scholarship (Academic Year 2023-2024)
  • 2024: 🥉 Third Prize, Academic Star (Academic Year 2023-2024)
  • 2023: 🏅 National Encouragement Scholarship (Academic Year 2022-2023)
  • 2023: 🥈 Second Prize, Academic Star (Academic Year 2022-2023)
  • 2022: 🥈 Second Prize, Academic Star (Academic Year 2021-2022)

Selected Publications

CMDS-AD: Cross-Modal Dual-Stream Decoupling for Few-Shot Anomaly Detection cover
2026

CMDS-AD: Cross-Modal Dual-Stream Decoupling for Few-Shot Anomaly Detection

Junhao Cai, Junyu Chen, Deyu Zeng#, Junhao Pang, Qiwei Liang, Xiaopin Zhong, Zongze Wu (# corresponding author)
European Conference on Computer Vision (ECCV), 2026
Paper Code Project
Cross-Modal Dynamic Hypergraph Computation via Functional-Structural Brain Network for Brain Disease Diagnosis cover
2026

Cross-Modal Dynamic Hypergraph Computation via Functional-Structural Brain Network for Brain Disease Diagnosis

Jingxi Feng, Heming Xu, Rundong Xue, Junhao Cai, Xudong Chen, Dong Zhang, Shaoyi Du
International Joint Conference on Artificial Intelligence (IJCAI), 2026
Beyond Viewpoint Generalization: What Multi-View Demonstrations Offer and How to Synthesize Them for Robot Manipulation? cover
2026

Beyond Viewpoint Generalization: What Multi-View Demonstrations Offer and How to Synthesize Them for Robot Manipulation?

Boyang Cai, Qiwei Liang, Jiawei Li, Shihang Weng, Zhaoxin Zhang, Tao Lin, Xiangyu Chen, Wenjie Zhang, Jiaqi Mao, Weisheng Xu, Bin Yang, Jiaming Liang, Junhao Cai, Renjing Xu
arXiv preprint, 2026
Paper Code Project
ForgeDreamer: Industrial Text-to-3D Generation with Multi-Expert LoRA and Cross-View Hypergraph cover
2026

ForgeDreamer: Industrial Text-to-3D Generation with Multi-Expert LoRA and Cross-View Hypergraph

Junhao Cai*, Deyu Zeng*, Junhao Pang, Lini Li, Zongze Wu, Xiaopin Zhong# (* equal contribution, # corresponding author)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR Findings), 2026
Paper Code Project
Cross-Modal Brain Graph Transformer via Function-Structure Connectivity Network for Brain Disease Diagnosis cover
2025

Cross-Modal Brain Graph Transformer via Function-Structure Connectivity Network for Brain Disease Diagnosis

Jingxi Feng, Heming Xu, Junhao Cai, Yujie Chang, Dong Zhang, Shaoyi Du, Juan Wang
Medical Image Computing and Computer Assisted Intervention (MICCAI), 2025
Paper