Efficient Multimodal Intelligence Laboratory

Home / People & Photos / Research / Publications / Teaching - Advanced Deep Learning

Course Descriptions

• This course introduces the core architectures, training methods, tuning strategies, and evaluation techniques of modern AI models, with a focus on Attention, Transformers, and Large Language Models (LLMs). It also covers recent extensions of LLMs, including reasoning, Multimodal Large Language Models (MLLMs), Agentic AI, and Retrieval-Augmented Generation (RAG). Students will gain basic hands-on experience through assignments using PyTorch and Hugging Face.

Textbook

• The course will primarily use lecture slides and selected reference materials, which may include:

  • CS224N: Natural Language Processing with Deep Learning (https://web.stanford.edu/class/archive/cs/cs224n/cs224n.1246)

  • CS231N: Deep Learning for Computer Vision (https://cs231n.stanford.edu/2025)

  • CME295: Transformers & Large Language Models (https://cme295.stanford.edu/)