About me

Hello! I’m Zhimu Zhou, a third-year undergraduate student from Beijing University of Posts and Telecommunications, majoring in Internet of Things Engineering. I am passionate about research in multimodal models, embodied AI, and human-robot interaction.

🔭 My long-term research goal: To develop intelligent systems that can seamlessly integrate multimodal information and interact with the physical world, ultimately creating more intuitive and effective human-robot collaboration frameworks.

Check out my detailed CV (English) / CV (中文)

🎯 Research Interests

• Multimodal Models
• Embodied AI
• Human-Robot Interaction
• Computer Vision
• Machine Learning

📚 Selected Publications

You can find the full list of my publications here.

• MSNav: Zero-Shot Vision-and-Language Navigation with Dynamic Memory and Feature Enhancement
  Chenghao Liu, Zhimu Zhou, Jiachen Zhang, Minghao Zhang, Songfang Huang, Huiling Duan.
  Under Review (*Co-first Author, randomly ordered by dice rolling)
  Preprint

• TTF-VLA: Temporal Token Fusion via Pixel-Attention Integration for Vision-Language-Action Models
  Chenghao Liu, Jiachen Zhang, Chengxuan Li, Zhimu Zhou, Songfang Huang, Huiling Duan.
  Under Review (Fourth Author)
  Preprint

• GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning
  Jian Zhao, Runze Liu, Kaiyan Zhang, Zhimu Zhou, Junqi Gao, Dong Li, Jiafei Lyu, Zhouyi Qian, Biqing Qi, Xiu Li, Bowen Zhou.
  arXiv preprint arXiv:2505.15825 (2025)
  Paper

🏆 Awards & Honors

• National Level: First Prize, IC Innovation Contest (MIIT 2024) - National No. 1
• Scholarships: Xiaomi Social Scholarship (Only 1 in Major, BUPT 2024), Second-Class Scholarship (Major Top 10%, BUPT 2023, 2024)
• Provincial Level: First Prize, China Innovation and Entrepreneurship Competition Beijing (Provincial Top 10%, Ministry of Education 2024), Second Prize, Computer Design Competition (Provincial Top 15%, Ministry of Education 2024)

🛠️ Research Projects

• MSNav: Zero-Shot Vision-and-Language Navigation with Dynamic Memory and Feature Enhancement (2025.01 - 2025.05)
  • Background: Traditional intelligent agents can only passively receive all information and cannot actively filter relevant information for tasks to form task memory
  • Contribution: Proposed MSNav, a modular framework with dynamic topological memory and spatial capabilities
  • Role: Co-first author, responsible for all experimental implementation and part of paper writing
  • Status: Under Review, Preprint
• TTF-VLA: Temporal Token Fusion via Pixel-Attention Integration for Vision-Language-Action Models (2025.05 - 2025.07)
  • Background: Need to enhance basic performance of VLA models without additional training costs
  • Contribution: Proposed visual feature reuse method, validated feasibility through experiments considering locality principle in VLA field
  • Role: Fourth author, focused on application of locality principle and experimental validation
  • Status: Under Review, Preprint
• VLA Strategy Guidance Based on the Process Reward Model (2024.09 - 2025.02)
  • Background: Inspired by verification effectiveness of NP problems
  • Contribution: Proposed independent process reward model to provide dense guidance for VLA models to avoid ineffective exploration
  • Role: Project leader, designed overall architecture and verified through virtual simulation experiments
• Chasing the Silver Bullet: An Inquiry into the Potential and Limits of Vibe-Coding (2025.08 - Present)
  • Background: “Silver Bullet” effect in software engineering - linear expressive growth of natural language cannot match non-linear growth of project complexity
  • Contribution: Independently designed and developed “Vibe Entropy” evaluation system to explore optimal theoretical boundaries of Vibe-Coding
  • Role: Project leader

💻 Technical Projects

• AI Driver Fatigue Detection System Based on Shaolin Pi (2024.03 - 2024.08)
  • Role: Program setup and model deployment leader
  • Contribution: Built Linux virtual development environment, developed CNN-based eye classification model, constructed 5k+ training dataset, deployed and tested on real devices
  • Achievement: Won National First Prize and Enterprise Award in National College Student Integrated Circuit Innovation and Entrepreneurship Competition

🎓 Education

• Beijing University of Posts and Telecommunications
  • Bachelor in Internet of Things Engineering (2022.09 - Present)
  • English Proficiency: CET-6: 627

Email / Github