发表机构
Beihang University; School of Economics and Management, Tsinghua University（北京航空航天大学; 清华大学经济管理学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对异质处理效应估计的局部不稳定性，提出CURL方法，通过LLM生成分离的分配与异质性导向表示，在四个基准上提升了10种主学习者的性能。
AI 中文摘要
估计异质处理效应是定向干预（如个性化推广、精准医疗）的核心。我们关注条件平均处理效应（CATE），即表征此类异质性的标准估计量。即便在标准识别条件下，有限样本CATE估计仍需学习协变量调整和处理效应异质性的干扰结构，通常还需结合协变量X的有效表示。原始数值与类别编码会使语义关系和高阶交互隐含，导致该联合任务局部不稳定。一项启发性研究进一步表明，这种不稳定性通过部分可分的分配侧与异质性侧通道显现。基于此，我们提出CURL（因果不确定性引导表示学习），这是一种插件适配器，利用估计器不确定性将预训练语义容量分配给局部不稳定单元。CURL通过两个角色条件提示查询冻结的大语言模型（LLM），从观测协变量构建分配导向与异质性导向的表示，并通过分离的路径传输。在四个基准测试中，CURL在多数设置下提升了10种主学习者的性能，而消融、细化动态、路径重分配及探测分析均支持该设计及两个通道的预期作用。
英文摘要
Estimating heterogeneous treatment effects is central to targeted interventions, such as personalized promotions and precision medicine. We focus on the conditional average treatment effect (CATE), a standard estimand for characterizing such heterogeneity. Even under standard identification conditions, finite-sample CATE estimation requires learning the nuisance structure for covariate adjustment and treatment-effect heterogeneity, often together with an effective representation of X. Raw numerical and categorical encodings can leave semantic relations and higher-order interactions implicit, making this joint task locally unstable. A motivating study further shows that this instability appears through partially separable assignment- and heterogeneity-side channels. Building on this observation, we propose CURL (Causal Uncertainty-guided Representation Learning), a plug-in adapter that uses estimator uncertainty to allocate pretrained semantic capacity to locally unstable units. CURL queries a frozen LLM through two role-conditioned prompts, constructs assignment- and heterogeneity-oriented representations from the observed covariates, and routes them through separated pathways. On four benchmarks, CURL improves ten host learners in most settings, while ablation, refinement-dynamics, route-reassignment, and probe analyses support the intended design and roles of the two channels.
Comments17 pages, 12 figures, 5 tables
