Cite
Notes
Only stored in your browser.
Attribution
MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding
arXiv 2024
from 1 papers
Avner May
Beidi Chen
Ian En-Hsu Yen
Jian Chen
Jinyuan Shi
Ranajoy Sadhukhan
Ruihang Lai
Tianqi Chen
Zhuoming Chen