CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing

arXiv:2305.11738v3 Announce Type: replace
Abstract: Recent developments in large language models (LLMs) have been impressive. However, these models sometimes show inconsistencies and problematic behavior, such as hallucinating facts, generating flawed code, or creating offensive and toxic content. Unlike these models, humans typically utilize external tools to cross-check and refine their initial content, like using a search engine for fact-checking, or a code interpreter for debugging. Inspired by this observation, we introduce a framework called CRITIC that allows LLMs, which are …

abstract arxiv behavior check code cs.ai cs.cl facts humans interactive language language models large language large language models llms refine show tool tools type