符号化思维链蒸馏：小模型也能“逐步思考”
Symbolic Chain-of-Thought Distillation: Small Models Can Also "Think" Step-by-Step
AI总结：
本文提出SCoTD，用大教师模型生成的多条推理链蒸馏125M至1.3B小学生模型，证明小模型也能通过逐步推理在常识任务上显著获益。
摘要：
思维链提示（例如“让我们一步一步地思考”）会促使大型语言模型为其预测表述推理过程。尽管思维链能够带来显著的性能提升，但其收益似乎只有在足够大的模型（超过50B参数）上才会显现。我们证明，参数规模小几个数量级的模型（125M至1.3B参数）仍然能够从思维链提示中受益。为实现这一点，我们提出了 Symbolic Chain-of-Thought Distillation（SCoTD，符号化思维链蒸馏），这是一种利用从显著更大的教师模型采样得到的推理过程来训练较小学生模型的方法。在多个常识基准上的实验表明：1）SCoTD在监督和少样本设置下均能提升学生模型的性能，尤其在挑战集上表现突出；2）为每个实例从教师模型采样多条推理链至关重要；3）蒸馏之后，尽管学生模型的参数少了几个数量级，人类评判认为其思维链与教师模型相当。我们还检验了若干关于思维链样本何种属性重要的假设，例如多样性、教师似然度与开放性之间的对比。我们公开发布了思维链样本语料库和代码。
英文摘要：
Chain-of-thought prompting (e.g., "Let's think step-by-step") primes large language models to verbalize rationalization for their predictions. While chain-of-thought can lead to dramatic performance gains, benefits appear to emerge only for sufficiently large models (beyond 50B parameters). We show that orders-of-magnitude smaller models (125M -- 1.3B parameters) can still benefit from chain-of-thought prompting. To achieve this, we introduce Symbolic Chain-of-Thought Distillation (SCoTD), a method to train a smaller student model on rationalizations sampled from a significantly larger teacher model. Experiments across several commonsense benchmarks show that: 1) SCoTD enhances the performance of the student model in both supervised and few-shot settings, and especially for challenge sets; 2) sampling many reasoning chains per instance from the teacher is paramount; and 3) after distillation, student chain-of-thoughts are judged by humans as comparable to the teacher, despite orders of magnitude fewer parameters. We test several hypotheses regarding what properties of chain-of-thought samples are important, e.g., diversity vs. teacher likelihood vs. open-endedness. We release our corpus of chain-of-thought samples and code.
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。
