Multi-expert Prompting Improves Reliability, Safety, and Usefulness of Large Language Models

Open in new window