CPEG 592XNatural Language Processing an

Catalog

Natural Language Processing (NLP) and its applications via Large Language Models (LLMs) has revolutionized the AI field. This course starts out with foundations of NLP and Transformer-based NLP architectures such as BERT and GPT. Then it elaborates on the different recent improvements in Transformer based designs such as RoBERTa, Transformer-XL, Linformer, Longformer, Long-Short Transformer, Perceiver. The more recent architectures for NLP based on State Space Models are also presented. Fine tuning of LLMs such as ChatGPT, Gemini, Llama via Low Rank Adaptations (LORA) and QLORA are explained and implemented as class projects. The course also covers important recent research papers in the NLP field.