Applied Natural Language Processing (COMP6965001)

Applied Natural Language Processing for the Artificial Intelligence program, centered on modern large language models: how language is represented as numbers, how transformers work, and how to build, retrieve, and fine-tune real NLP systems.

Sessions

• Sessions 1-3: Language Models — introduction to language models, tokens and embeddings, looking inside transformer LLMs
• Sessions 4-6: Working with Text — text classification, text clustering and topic modeling, prompt engineering
• Sessions 7-9: Generation and Search — advanced text generation techniques and tools, semantic search and retrieval-augmented generation, multimodal large language models
• Sessions 10-12: Building and Fine-tuning — creating text embedding models, fine-tuning representation models for classification, fine-tuning generation models
• Session 13: Review and Presentation

Resources

• Supplementary: Alammar and Grootendorst. Hands-On Large Language Models: Language Understanding and Generation (O’Reilly)
• Tools: Hugging Face Transformers, PyTorch, Google Colab

Full session materials and lab guides are shared with enrolled students only.