Fast Chain-of-Thought: A Glance of Future from Parallel Decoding Leads to Answers Faster
CoRR(2023)
摘要
In this work, we propose FastCoT, a model-agnostic framework based onparallel decoding without any further training of an auxiliary model ormodification to the LLM itself. FastCoT uses a size-varying context windowwhose size changes with position to conduct parallel decoding andauto-regressive decoding simultaneously, thus fully utilizing GPU computationresources. In FastCoT, the parallel decoding part provides the LLM with a quickglance of the future composed of approximate tokens, which could lead to fasteranswers compared to regular autoregressive decoding used by causaltransformers. We also provide an implementation of parallel decoding withinLLM, which supports KV-cache generation and batch processing. Through extensiveexperiments, we demonstrate that FastCoT saves inference time by nearly 20with only a negligible performance drop compared to the regular approach.Additionally, we show that the context window size exhibits considerablerobustness for different tasks.
更多查看译文
关键词
Meta-Learning
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn