CS 5624: Natural Language Processing

Course description

This course provides an in-depth exploration of Natural Language Processing (NLP) with a focus on the principles and techniques behind the development of large language models (LLMs) and LLM agents. Topics include foundational NLP concepts, neural language models, Transformers architectures, LLM pretraining and post-training, efficient training and inference, LLM reasoning and test-time scaling, interpretability and evaluation, multimodal and diffusion LLMs, AI agents, recursive self-improvement, harness evolution and continual learning, automated scientific discovery, and AI security and privacy, etc. Students will gain practical experience through coding assignments and a final project, which prepare them to apply state-of-the-art models and techniques to real-world applications.

Prerequisites No prerequisites are required for this course; however, the following could be helpful:

• Familiarity with basic machine learning concepts
• Familiarity with basic statistical concepts
• Familiarity with Python or coding agents

---

• Time: Tuesday & Thursday 2:00 - 3:15 PM
• Location: Fralin Hall 102 (in person) / Zoom (online)
• Contact: Students should ask all course-related questions on Piazza or Discord. For personal matters or emergencies, please email the instructor at cs5624instructors@gmail.com. For anonymous questions or comments, please use this form.

---