Large language models are increasingly used as proxies for human subjects in social science research, yet external validity requires that synthetic agents faithfully reflect the preferences of target human populations. We introduce *preference reconstruction theory*, a framework that formalizes preference alignment as a representation learning problem: constructing a functional basis of proxy agents and recovering population preferences through weighted aggregation. We implement this via *Prompts to Proxies* ($\texttt{P2P}$), a modular two-stage system. Stage 1 uses structured prompting with entropy-based adaptive sampling to construct a diverse agent pool spanning the latent preference space. Stage 2 employs L1-regularized regression to select a compact ensemble whose aggregate response distributions align with observed data from the target population. $\texttt{P2P}$ requires no finetuning and no access to sensitive demographic data, incurring only API inference costs. We validate the approach on 14 waves of the American Trends Panel, achieving an average test MSE of 0.014 across diverse topics at approximately 0.8 USD per survey. We additionally test it on the World Values Survey, demonstrating its potential to generalize across locales. When stress-tested against an SFT-aligned baseline, $\texttt{P2P}$ achieves competitive performance using less than 3% of the training data.

翻译：大型语言模型日益成为社会科学研究中人类受试者的代理，但外部效度要求合成智能体能够忠实反映目标人群的偏好。我们提出*偏好重构理论*，该框架将偏好对齐形式化为表示学习问题：构建代理智能体的函数基，并通过加权聚合还原群体偏好。我们通过*从提示到代理*（$\texttt{P2P}$）系统实现该理论——这是一个模块化的两阶段系统。第一阶段采用基于熵的自适应采样结构化提示技术，构建覆盖潜在偏好空间的多样化智能体池。第二阶段使用L1正则化回归选择紧凑型集成，其聚合响应分布与目标人群的观测数据对齐。$\texttt{P2P}$无需微调模型且不涉及敏感人口统计数据，仅产生API推理成本。我们在美国趋势面板的14轮调查数据上验证该方法，在约0.8美元/次调查的成本下，跨多元主题实现平均测试均方误差0.014。我们进一步在世界价值观调查数据上进行测试，证明其跨地域泛化潜力。在与监督微调对齐基线的压力测试中，$\texttt{P2P}$使用不足3%的训练数据即达到可比性能。