#natural language process...
Jun 2026
JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting
JetSpec is proposed, a head-based SD framework that combines one-forward drafting efficiency with branch-wise causal conditioning that enables JetSpec to convert larger draft budgets into longer accepted prefixes and higher end-to-end speedup.
Lan-Xiang Hu, Zhao-Xiang Feng, Yulun Wu et al.
· arXiv.org · 15 citations
· ⚡1