arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2512.20162cs.AI
收藏

人类与大语言模型的概念泛化:来自数字游戏的洞察

Concept Generalization in Humans and Large Language Models: Insights from the Number Game

Department of Biomedical Engineering, Sungkyunkwan University, South Korea(生物医学工程系,全州大学,韩国)

Arghavan Bazigaran, Hansem Sohn

AI总结:

研究通过数字游戏对比人类与大语言模型在概念泛化上的差异,揭示人类灵活的推理策略与LLM依赖数学规则的本质区别。

摘要:

我们比较了人类和大语言模型(LLM)在数字游戏中的泛化能力,这是一种概念推理任务。使用贝叶斯模型作为分析框架,我们研究了人类和LLM的归纳偏置和推理策略。贝叶斯模型在人类行为上表现更优,因为人类能够灵活地推断基于规则和相似性的概念,而LLM更依赖数学规则。人类还展示了少量样本的泛化能力,即使仅从一个例子中也能进行泛化,而LLM则需要更多样本才能泛化。这些对比突显了人类和LLM在数学概念推理和泛化方面的根本差异。

英文摘要:

We compare human and large language model (LLM) generalization in the number game, a concept inference task. Using a Bayesian model as an analytical framework, we examined the inductive biases and inference strategies of humans and LLMs. The Bayesian model captured human behavior better than LLMs in that humans flexibly infer rule-based and similarity-based concepts, whereas LLMs rely more on mathematical rules. Humans also demonstrated a few-shot generalization, even from a single example, while LLMs required more samples to generalize. These contrasts highlight the fundamental differences in how humans and LLMs infer and generalize mathematical concepts.

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。

↑