7 citations · 18 across the 9 of their papers we have counts for
1 paper · 1 filter
Roy Eisenstadt, Ido Cohen, Edo Cohen-Karlik +2
Speculative Decoding (SD) has significantly accelerated Large Language Model (LLM) inference, yet existing approaches face a fundamental tradeoff between two drafting strategies: n…