Skip to content
View Theia-4869's full-sized avatar
💻
💻
41 followers · 5 following

Achievements

Achievement: Starstruck

Achievements

Achievement: Starstruck

Highlights

  • Pro

Block or report Theia-4869

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Theia-4869/README.md

👤 About Me

I am currently a Ph.D. candidate at HMI Lab, NERCV²T, School of Computer Science, Peking University, supervised by Prof. Shanghang Zhang. I received my Bachelor’s degree in Artificial Intelligence (Turing Honor Degree) from Peking University in 2023, where I also obtained a Bachelor’s degree in Economics.

🔭 Research Interests

My research interests lie in multimodal large language models, including visual foundation models, vision language models, unified multimodal models, visual complex reasoning, visual model efficiency, and visual continual learning. The overall goal of my research is to develop a large-scale efficient visual perception system with human-like expression, adaptation, and generalization, equipped with powerful abilities including fundamental perception, cognitive reasoning, and autonomous creativity.

📬 Contact

📧 Email: theia@pku.edu.cn, theia4869@gmail.com

Feel free to reach out for collaboration!

Pinned Loading

  1. FasterVLM FasterVLM Public

    Official code for paper: [CLS] Attention is All You Need for Training-Free Visual Token Pruning: Make VLM Inference Faster.

    Python 114 4

  2. VisPruner VisPruner Public

    [ICCV 2025] Official code for paper: Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs

    Python 83 8

  3. CDPruner CDPruner Public

    [NeurIPS 2025] Official code for paper: Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs.

    Python 104 5

  4. Time-Search/TimeSearch-R Time-Search/TimeSearch-R Public

    [ICLR 2026] Official code for paper: TimeSearch-R: Adaptive Temporal Search for Long-Form Video Understanding via Self-Verification Reinforcement Learning.

    Python 27 8