Closed dingjingzhen closed 2 weeks ago
Hi @dingjingzhen ,
Thanks for your interest in our work. While Quest focus on boosting the efficiency of the decoding phase, the efficiency of prefill phase under long-context case is indeed a great issue, as explored by recent works like MInference. We believe the idea of query-aware sparsity can be easily extended to prefill phase for the following reasons:
Great job,but It seems that this article does not speed up the prefill phase, which in many cases takes a long time in the reasoning process of long text