Improving Reasoning in Large Language Models through General-Purpose Verification of Chain-of-Thought Prompting

Leveraging general-purpose verifiers to improve the accuracy and consistency of reasoning steps generated by large language models through chain-of-thought prompting.