It Takes Two to Tango: Accelerating Code Generation via Cache-Inspired Hierarchical Speculative Decoding

This repository contains code and the distilled checkpoints of the draft models for CodeLlama-34b and Deepseek-Code-33b for the paper “It Takes Two to Tango: Accelerating Code Generation via Cache-Inspired Hierarchical Speculative Decoding”. CacheCoder.zip contains the source code of this research. checkpoints.zip contains the distilled weights of the draft model used in this research.本代码仓库收录了论文"It Takes Two to Tango：基于缓存启发的分层投机解码（Speculative Decoding）加速代码生成"中，针对CodeLlama-34b与Deepseek-Code-33b所使用的草稿模型的代码与蒸馏模型检查点。其中，CacheCoder.zip 包含本研究的全部源代码，checkpoints.zip 则收录了本研究中所用草稿模型的蒸馏权重。