Search papers, labs, and topics across Lattice.
This paper introduces CURL (Causal Uncertainty-guided Representation Learning), a novel approach that leverages estimator uncertainty to enhance the estimation of conditional average treatment effects (CATE) in heterogeneous treatment effect scenarios. By querying a frozen large language model (LLM) with role-conditioned prompts, CURL constructs specialized representations for assignment and treatment-effect heterogeneity, addressing the instability caused by traditional encoding methods. The method demonstrates significant improvements across four benchmarks, outperforming ten host learners in various settings, thereby validating its design and effectiveness in enhancing treatment effect estimation.
CURL effectively harnesses LLMs to stabilize CATE estimation, leading to improved performance in personalized interventions across multiple benchmarks.
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.
Tsinghua AI