Now, I am a 2nd year master at Tsinghua University (M.Eng.@THU’2026), supervised by Prof. Yansong Tang. I obtained my bachelor's degree from the School of Software Engineering at Tongji University in 2023.
- 🔭 I’m currently working on large vision language models!
- 💬 How to reach me: Email.
- 📫 Recent work:
- ATP-LLaVA. [Under Review] We propose ATP-LLaVA, a framework that adaptively determines pruning ratios instance-wise and LLM layer-wise for effective visual token pruning on large vision language models.
- VoCo-LLaMA. [Under Review] The first approach to compress vision information utilizing the LLMs' understanding paradigm, which can compress hundreds of vision tokens into a single VoCo token with minimal visual information loss.
- LAVT-RS. [CVPR2022, TPAMI2024] Pixel-level language-aware early-fusion vision transformer structure for both referring image and video segmentation.