← 返回卡片

原书出处

Designing Large Language Model Applications

First Edition · 核验版本的摘录快照 · 导入于 2026-11-14T13:54:13.944Z

Speculative Decoding The concept behind speculative decoding is simple. A smaller model, called a draft model, is used to generate several subsequent candidate output tokens. Then, the main larger model is used to compute the conditional probabilities of the candidate output tokens at once, using them to decide which tokens to accept and which ones to reject. The more draft tokens accepted, the better the

此预览只包含定位所需的摘录和上下文。

原书定位详情

来源:a19c7933-4eda-4176-aebd-30fe7e1ad292

正文位置:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch09.html

EPUB SHA-256:1b4d2347fec2f949fcddfbaf4fad597a1ae20887a096cfae64c59a619b4781bf