通过情境形成与导航提升LLM社交模拟中的行为一致性
Improving Behavioral Alignment in LLM Social Simulations via Context Formation and Navigation
AI总结：
本文提出通过情境形成与导航提升LLM在复杂决策环境中的行为一致性，验证了两阶段框架在不同任务中的有效性。
摘要：
大型语言模型（LLM）越来越多地被用于实验环境中模拟人类行为，但在复杂决策环境中，它们系统性地偏离人类决策，因为参与者必须预测他人的行动并基于观察到的行为形成信念。我们提出了一种两阶段框架以改进行为一致性。第一阶段，情境形成，明确指定实验设计以建立准确的决策任务和情境的表示。第二阶段，情境导航，引导在该表示中的推理过程以做出决策。我们通过一个顺序购买游戏的焦点复制（Kremer和Debo，2016）验证了该框架，并扩展到一个具有成本信号的众筹游戏（Cason等，2025）和一个需求估计任务（Gui和Toubia，2025）以测试在不同决策环境中的一般性。在四个最先进的（SOTA）模型（GPT-4o、GPT-5、Claude-4.0-Sonnet-Thinking、DeepSeek-R1）上，我们发现复杂决策环境需要两个阶段才能与人类基准实现行为一致性，而简单的需求估计任务仅需情境形成。我们的发现澄清了每个阶段何时是必要的，并为设计和诊断LLM社交模拟提供了系统的方法，作为人类受试者在行为研究中的补充。
英文摘要：
Large language models (LLMs) are increasingly used to simulate human behavior in experimental settings, but they systematically diverge from human decisions in complex decision-making environments, where participants must anticipate others' actions and form beliefs based on observed behavior. We propose a two-stage framework for improving behavioral alignment. The first stage, context formation, explicitly specifies the experimental design to establish an accurate representation of the decision task and its context. The second stage, context navigation, guides the reasoning process within that representation to make decisions. We validate this framework through a focal replication of a sequential purchasing game with quality signaling (Kremer and Debo, 2016), extending to a crowdfunding game with costly signaling (Cason et al., 2025) and a demand-estimation task (Gui and Toubia, 2025) to test generalizability across decision environments. Across four state-of-the-art (SOTA) models (GPT-4o, GPT-5, Claude-4.0-Sonnet-Thinking, DeepSeek-R1), we find that complex decision-making environments require both stages to achieve behavioral alignment with human benchmarks, whereas the simpler demand-estimation task requires only context formation. Our findings clarify when each stage is necessary and provide a systematic approach for designing and diagnosing LLM social simulations as complements to human subjects in behavioral research.
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。
