自动生成并优化推理步骤，提升大模型少样本学习效果
Augmenting In-Context-Learning in LLMs via Automatic Data Labeling and Refinement
- 从少量人工示例出发，自动构造带中间步骤的演示数据
- 在代码问答和数学推理任务中最高提升5.5%准确率
- 适合需要高质量推理示范的AI系统开发者
大型语言模型在多种任务上可通过思维链（CoT）或上下文学习（ICL）提升性能，这类方法依赖包含求解步骤的示例。然而，编写包含中间步骤的示范数据需大量人工投入，尤其当步骤涉及可执行程序或逐步推理时。本文提出自动数据标注与优化（ADLR）方法，从少量人工设计的种子示例出发，自动生成并筛选包含中间步骤的演示数据。我们在基于代码的表格问答和数学推理任务中验证了该方法的有效性，最高实现5.5%的性能提升。论文补充材料提供了完整代码实现，并将公开发布。
原文摘要 · Abstract (English)
It has been shown that Large Language Models' (LLMs) performance can be improved for many tasks using Chain of Thought (CoT) or In-Context Learning (ICL), which involve demonstrating the steps needed to solve a task using a few examples. However, while datasets with input-output pairs are relatively easy to produce, providing demonstrations which include intermediate steps requires cumbersome manual work. These steps may be executable programs, as in agentic flows, or step-by-step reasoning as in CoT. In this work, we propose Automatic Data Labeling and Refinement (ADLR), a method to automatically generate and filter demonstrations which include the above intermediate steps, starting from a small seed of manually crafted examples. We demonstrate the advantage of ADLR in code-based table QA and mathematical reasoning, achieving up to a 5.5% gain. The code implementing our method is provided in the Supplementary material and will be made available.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。