Can Large Language Models Explore Meaning