Abstract:Autoregressive decoding is bottlenecked by its sequential nature. Speculative decoding has become a standard way to accelerate inference by using a fast draft model to predict upcoming tokens from a slower target model, and then verifying them in parallel with a single target model forward pass. However, speculative decoding itself relies on a sequential dependence between speculation and verification. We introduce speculative speculative decoding (SSD) to parallelize these operations. While a verification is ongoing, the draft model predicts likely verification outcomes and prepares speculations pre-emptively for them. If the actual verification outcome is then in the predicted set, a speculation can be returned immediately, eliminating drafting overhead entirely. We identify three key challenges presented by speculative speculative decoding, and suggest principled methods to solve each. The result is Saguaro, an optimized SSD algorithm. Our implementation is up to 2x faster than optimized speculative decoding baselines and up to 5x faster than autoregressive decoding with open source inference engines.
Sign up for the Breaking News US email to get newsletter alerts in your inbox，推荐阅读一键获取谷歌浏览器下载获取更多信息
。体育直播对此有专业解读
Let’s start by defining a couple types. First, our value type:
Зеленский решил отправить военных на Ближний Восток20:58。关于这个话题，旺商聊官方下载提供了深入分析
Friedmann, outside a Nashville courthouse in 2008.Photograph by Mark Humphrey / AP