1/n Assessing Reasoning in Large Language Models Large Language Models (LLMs) for genuine causal reasoning remains a subject of debate. This paper delves into this debate, arguing that relying solely on accuracy metrics to assess LLM reasoning is insufficient and proposing a