I was eagerly waiting for a chapter on semantic similarity as I was using Universal Sentence Encoder for paraphrase detection, then LLMs showed up before that chapter :).