Tag
Shared a prompt methodology based on LLM self-correction research, emphasizing that self-checking is limited without external feedback, and recommending progressively enhanced prompting strategies such as adversarial review.