[HN Gopher] Break the Sequential Dependency of LLM Inference Usi...
___________________________________________________________________
Break the Sequential Dependency of LLM Inference Using Lookahead
Decoding
Author : zhisbug
Score : 7 points
Date : 2023-11-21 20:12 UTC (2 hours ago)
(HTM) web link (lmsys.org)
(TXT) w3m dump (lmsys.org)
| zhisbug wrote:
| New parallel decoding algorithm that trades flops for latency
| reduction
___________________________________________________________________
(page generated 2023-11-21 23:03 UTC)