A [[large language model]] (LLM), in the context of [[natural language]] processing and [[artificial intelligence]], refers to a sophisticated [[neural network]] that has been trained on a massive amount of [[text]] [[data]] to understand and generate human-like language. These models are typically built on [[architecture]]s like [[transformer]]s.