Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs

Open in new window