https://openreview.net/pdf?id=RAtrnAtAsM2. methodology(1) Fine-grained World Knowledge Integration Module(2) Iterative Feedback and Refinement Moduleone LLM serve as Explainer - initial 결과물 생성Critic LLM - Obeservation, Iterative feed back 받음 -> refine its explanation 2.1 Fine-grained World Knowledge Integrationinception prompt -> 에 의해Cause Analyst role Effect Analyst role가 two LLMs에게 assign됨 The..