No matter how amazing large language models are, by design they may not be the best suited to serve as reliable knowledge models.

无论大语言模型有多牛，它们可能本来就不是可靠的知识模型的最佳选择。