I am a master’s student at Harbin Institute of Technology (Shenzhen). Prior to this, I obtained my bachelor’s degree from the same institution.
🤔 Research Interests
- Multimodal: Multimodal Large Language Models
- Efficient Training and Inference: Visual Token Reduction, Continual Learning, Model Compression
🔥 News
- 2026.02: One paper on efficient training for MLLMs is out on arXiv, see here.
- 2025.10: Honored to join LONG Group at HKUST as a visiting student, working with Prof. Long Chen.
- 2025.08: One paper on continual learning for MLLMs is accepted by EMNLP 2025 main, see here.
- 2025.08: One Zhihu blog on Mechanism of MLLMs is forwarded by PaperWeekly.
- 2024.04: One paper on compressing diffusion models is out on arXiv, see here.
- 2023.09: Excited to be a Research Intern at OPPO AI Center.
📝 Publications
arXiv
Fast-Slow Efficient Training for Multimodal Large Language Models via Visual Token Pruning
Dingkun Zhang, Shuhan Qi, Yulin Wu, Xinyu Xiao, Xuan Wang, Long Chen
- A framework to accelerate MLLMs training via visual token pruning.
- Overcome the training-inference mismatch when introducing visual token pruning to training time.
EMNLP 2025 main
Merge then Realign: Simple and Effective Modality-Incremental Continual Learning for Multimodal LLMs
Dingkun Zhang, Shuhan Qi, Xinyu Xiao, Kehai Chen, Xuan Wang
- Extend existing MLLMs to more modalities efficiently.
- Minimalist, Anti-Catastrophic Forgetting.
arXiv
LAPTOP-Diff: Layer Pruning and Normalized Distillation for Compressing Diffusion Models
Dingkun Zhang*, Sijia Li*, Chen Chen, Qingsong Xie, Haonan Lu
- Compress diffusion models through layer pruning and knowledge distillation.
📖 Educations
- 2024.06 - present, Master’s Student, Harbin Institute of Technology (Shenzhen).
- 2020.09 - 2024.06, Undergraduate Student, Harbin Institute of Technology (Shenzhen).
💻 Internships
- 2025.10 - 2026.02, Visiting Student, Hong Kong University of Science and Technology.
- 2023.09 - 2024.04, Research Intern, OPPO AI Center.