Cache-to-Cache: Direct Semantic Communication Between Large Language Models(arxiv.org)
13 points by jonbaer 2 days ago | discuss