Small Language Models Need Strong Verifiers to Self-Correct Reasoning

Open in new window