1 paper · 1 filter
Yue Guan, Changming Yu, Shihan Fang +8
Speculative decoding improves LLM inference by generating and verifying multiple tokens in parallel, but existing systems suffer from suboptimal performance due to a mismatch betwe…