35 citations
2 papers
cs.CL2023★ 35 cited
Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding
Jun Zhang, Jue Wang, Huan Li +4
We present a novel inference scheme, self-speculative decoding, for accelerating Large Language Models (LLMs) without the need for an auxiliary model. This approach is characterize…
cond-mat.mtrl-sci2017
Quantifying the structural integrity of nanorod arrays
Florian Thöle, Longjian Xue, Claudia Heß +3
Arrays of aligned nanorods oriented perpendicular to a support, which are accessible by top-down lithography or by means of shape-defining hard templates, have received increasing…