ZJUNICE Tutorial 2025

Posted on Edited on In

This year, my team released a series of video tutorials on Bilibili, covering topics ranging from generative models (e.g., large language models and diffusion models) to reinforcement learning, as well as their intersection and applications. We hope this series provides a clear explanation of the motivations behind our research and helps inspire potential collaborations. We also aim for these tutorials to serve as a valuable resource for prospective students who are interested in understanding what we do and why we do it.

ZJUNICE Tutorial 2025
├─ Generative Models
│ ├─ Large Language Models (LLMs)
│ │ ├─ Basics
│ │ │ └─ How Can We Use LLMs: A Brief Tutorial (, )
│ │ │ └─ Architectural Innovation and Implementation Details of LLMs (, )
│ │ │ └─ A Technical Analysis of Multimodal Large Language Models (, )
│ │ └ └─ AI Agent and Agentic Workflow based in Large Language Model (, )
│ │ └─ Applications
│ │ │ └─ LLMs for Next-Generation Networks: An AI-Native CP Perspective (, )
│ │ └ └─ Large Models as Compressors for Efficient Communication (, )
│ └─ Generative Models
│ │ ├─ Basics
│ └ └ └─ Advanced Generative Models: From VAE to Diffusion and Flow Matching (, VAE: , DM: , FM: )
├─ Reinforcement Learning
│ │ ├─ Basics
│ │ │ └─ The Evolution of RL: From Single-Agent to Multi-Agent Paradigms (, )
│ │ │ └─ Model-based Reinforcement Learning (MBRL) (, )
│ └ └ └─ Towards General-Purpose Reinforcement Learning Agents (, )
└── Intersections
│ ├─ Basics
│ │ └─ Evolution and Application of Diffusion Models in RL (, )
│ └ └─ Evolution and Application of RL in LLMs (, )
│ ├─ Applications
└ └ └─ V2X Collaborative Perception (, )