让语言模型在编程求解中自我进化，提升代码生成能力。
Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI
- 将语言模型嵌入自迭代演化循环，交替进行搜索与回溯学习。
- 在ARC-AGI上实现52%的测试集求解率，模型规模越大效果越显著。
- 适合研究程序合成、大模型自我优化及进化式算法的开发者。
许多编程求解任务对当前最先进的语言模型而言仍过于复杂，单次尝试难以解决。基于搜索的演化方法通过迭代探索解空间提供了一种有前景的替代方案，但其效果受限于底层生成模型的固定能力。我们提出SOAR，一种将语言模型融入自改进演化循环的方法，以实现程序合成的持续优化。SOAR在(1)利用大语言模型采样和优化候选解的演化搜索，与(2)将搜索过程转化为有效问题-解对，用于微调语言模型采样与优化能力的回溯学习之间交替进行，从而在后续迭代中实现更高效的搜索。在具有挑战性的ARC-AGI基准上，SOAR在不同模型规模和迭代次数下均取得显著性能提升，并实现了采样与优化任务间的正向迁移。这些改进可推广至测试时适应，使SOAR成功解决52%的公开测试集题目。代码已开源：https://github.com/flowersteam/SOAR
原文摘要 · Abstract (English)
Many program synthesis tasks prove too challenging for even state-of-the-art language models to solve in single attempts. Search-based evolutionary methods offer a promising alternative by exploring solution spaces iteratively, but their effectiveness remain limited by the fixed capabilities of the underlying generative model. We propose SOAR, a method that learns program synthesis by integrating language models into a self-improving evolutionary loop. SOAR alternates between (1) an evolutionary search that uses an LLM to sample and refine candidate solutions, and (2) a hindsight learning phase that converts search attempts into valid problem-solution pairs used to fine-tune the LLM's sampling and refinement capabilities\, -- \,enabling increasingly effective search in subsequent iterations. On the challenging ARC-AGI benchmark, SOAR achieves significant performance gains across model scales and iterations, leveraging positive transfer between the sampling and refinement finetuning tasks. These improvements carry over to test-time adaptation, enabling SOAR to solve 52\% of the public test set. Our code is open-sourced at: https://github.com/flowersteam/SOAR
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。