基于语言与代码的概率推理实现归纳与探究

AI 摘要

该研究提出一种计算模型，将符号知识编码为结合自然语言与源代码的“心智程序”，并利用LLM引导的贝叶斯学习算法进行序列推断，以模拟人类的归纳学习与主动探究。实验表明，该模型能复现人类行为中的锚定、花园路径等效应，而纯LLM或经典贝叶斯模型则无法同时满足数据效率、不确定性表达和灵活性。研究推测，人类通过混合语言与程序表征的假设空间进行近似贝叶斯更新，LLM作为自底向上的神经机制使推理可行且可学习。

正文节选

Induction and Inquiry via Probabilistic Reasoning over Language and Code Abstract How humans grow and maintain abstract knowledge from the sparse, streaming noisy data of experience is a longstanding challenge in cognitive science. Any computational account must satisfy at least three desiderata: It must be (1) data-efficient and compute-efficient, (2) capture gradations of uncertainty to support intelligent inquiry and information gathering, and (3) be flexible enough to mentally represent the

发布时间：2026-09-03 12:00

抓取时间：2026-09-03 12:10

来源机构：arXiv