用结构化方法让大模型成功解决600个程序合成难题，揭示人机协作新思路。
Structured Program Synthesis using LLMs: Results and Insights from the IPARC Challenge
- 设计结构化归纳编程框架，结合大模型生成与人类修正。
- 在全部600个任务上达成成功解，突破以往自动化方案瓶颈。
- 揭示代码复用、冻结正确部分等关键协作机制，适合复杂编程研究者。
IPARC挑战赛受ARC启发，提供基于合成图像的程序合成任务，聚焦序列、选择和循环结构，共600个任务，此前难以自动求解。本文提出一种结合大模型的结构化归纳编程方法，成功解决了所有类别任务。其受控特性揭示了大模型代码生成的关键洞察：先验结构化的重要性、大模型辅助构建结构（需人工修正）、正确代码需冻结、代码复用效率高，以及生成代码可激发人类创造力。这些发现为复杂程序合成中的人机协作提供了有效机制。
原文摘要 · Abstract (English)
The IPARC Challenge, inspired by ARC, provides controlled program synthesis tasks over synthetic images to evaluate automatic program construction, focusing on sequence, selection, and iteration. This set of 600 tasks has resisted automated solutions. This paper presents a structured inductive programming approach with LLMs that successfully solves tasks across all IPARC categories. The controlled nature of IPARC reveals insights into LLM-based code generation, including the importance of prior structuring, LLMs' ability to aid structuring (requiring human refinement), the need to freeze correct code, the efficiency of code reuse, and how LLM-generated code can spark human creativity. These findings suggest valuable mechanisms for human-LLM collaboration in tackling complex program synthesis.
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。