决策模型作为可扩展的经济替代模型,捕捉人类行为差异
Decision Models as Scalable Economic Surrogates that Capture Human Variation
Maxime Saxena · Fabio Caccioli · R. Maria del Rio-Chanona
中文摘要
决策模型(Decision Models, DMs)是一类新型 AI 模型,能快速响应并返回固定选项集上的概率分布。这些特性使其成为经济模拟和调查中人类替代模型的候选,而这一角色正越来越多地由大语言模型(LLM)承担,但 LLM 在大规模应用时成本高昂且低估了人类行为的差异。首次评估 DMs 作为经济学中人类替代模型的表现,在三种经济场景下比较 10 个 DMs 与 8 个 LLMs:消费支出调查、若干经典经济博弈和劳动力市场田野实验。在通用决策基准上得分最高的前沿 DMs 在三种场景中均比 LLMs 更贴近地复现人类响应的差异。在支出调查中,它们捕捉人类众数响应的能力与 LLMs 相当,但在经济博弈中较弱。前沿 DMs 的速度约为最快 LLM 的十倍,成本可低至百分之一以下。这些结果表明,DMs 比 LLMs 更好地捕捉人类行为的异质性,并使大规模经济模拟更可行。
关键要点
- 01问题:LLM 作为经济模拟中的人类替代模型成本高且低估人类行为差异
- 02方法:在消费调查、经典经济博弈和劳动力市场田野实验中比较 10 个 DMs 与 8 个 LLMs
- 03结果:前沿 DMs 在三种场景中均比 LLMs 更贴近复现人类响应差异
- 04结果:前沿 DMs 速度约为最快 LLM 的十倍,成本可低至百分之一以下
- 05局限:在经济博弈中捕捉人类众数响应的能力弱于 LLMs
解读
尚无解读。
原始英文摘要
arXiv:2611.05553v1 Announce Type: cross Abstract: Decision Models (DMs), a new class of AI models, give fast responses and return a probability distribution over a fixed set of options. These properties make them candidate human surrogates in economic simulations and surveys, a role increasingly filled by Large Language Models (LLMs), which are costly at scale and under-represent the variation in human behaviour. We present a first evaluation of DMs as human surrogates in economics, comparing 10 DMs with 8 LLMs in three economic settings: a household spending survey on consumption, several classic economic games and a labour-market field experiment. Frontier DMs, those that score highest on general decision-making benchmarks, reproduce the variation in human responses more closely than LLMs in all three settings. They capture the modal human response as well as LLMs in the spending survey, but less well in the economic games. Frontier DMs are also roughly ten times faster than the fastest LLM and can be over a hundred times cheaper. Together, these results suggest that DMs capture the heterogeneity of human behaviour better than LLMs and make large-scale economic simulations more feasible.