Uncertainty, Prompting, and Chain-of-Thoughts in Large Language Models - Part 2

Learn about advanced concepts in AI uncertainty quantification and prompting techniques in this comprehensive lecture. Explore temperature scaling methods and Bayesian approaches to calibration before diving into free-text explanations and chain-of-thought prompting. Master in-context learning (ICL) principles and their reliable implementation, while understanding prompt-based fine-tuning strategies. Examine practical applications through case studies of FLAN-T5 and LLaMA Chat models. Gain insights into how these techniques improve AI model performance and reliability through detailed explanations and real-world examples.