CODECRASH: Exposing LLMFragility to Misleading Natural Language in Code Reasoning

Open in new window