思维链在从大型语言模型中蒸馏推理能力方面的有效性
Effectiveness of Chain-of-Thought in Distilling Reasoning Capability from Large Language Models
AI总结：
本文研究思维链在白盒知识蒸馏中的作用，利用Qwen和Llama2系列模型及CoT-Collection数据蒸馏较小LLM，并在BBH基准上验证其能提升自然语言推理与理解任务的平均性能。
摘要：
思维链（CoT）提示是一种被广泛用于提升大型语言模型（LLMs）推理能力的方法。近年来，CoT已被用于知识蒸馏（KD），以将推理能力从较大的LLM迁移到较小的LLM。本文研究了CoT在使用白盒知识蒸馏将推理能力从较大LLM蒸馏到较小LLM过程中的作用，分析了其在多种自然语言推理与理解任务中提升蒸馏模型性能的有效性。我们使用Qwen和Llama2系列LLM进行白盒知识蒸馏实验，并采用CoT-Collection数据集中的CoT数据。随后，在BIG-Bench-Hard（BBH）基准的自然语言推理与理解任务上对蒸馏模型进行评估，该基准对较小LLM构成了复杂挑战。实验结果表明，CoT在提升白盒知识蒸馏有效性方面发挥了作用，使蒸馏模型能够在BBH的自然语言推理与理解任务中取得更好的平均性能。
英文摘要：
Chain-of-Thought (CoT) prompting is a widely used method to improve the reasoning capability of Large Language Models (LLMs). More recently, CoT has been leveraged in Knowledge Distillation (KD) to transfer reasoning capability from a larger LLM to a smaller one. This paper examines the role of CoT in distilling the reasoning capability from larger LLMs to smaller LLMs using white-box KD, analysing its effectiveness in improving the performance of the distilled models for various natural language reasoning and understanding tasks. We conduct white-box KD experiments using LLMs from the Qwen and Llama2 families, employing CoT data from the CoT-Collection dataset. The distilled models are then evaluated on natural language reasoning and understanding tasks from the BIG-Bench-Hard (BBH) benchmark, which presents complex challenges for smaller LLMs. Experimental results demonstrate the role of CoT in improving white-box KD effectiveness, enabling the distilled models to achieve better average performance in natural language reasoning and understanding tasks from BBH.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
