[EMNLP 2024] Multi-expert Prompting Improves Reliability, Safety and Usefulness of Large Language Models - View it on GitHub