2 ms·Cache-to-Cache: Direct Semantic Communication Between Large Language Models1 points by mooreds 1y ago