arXiv cs.AI· Wenxin Chen, Kyra Gan, Mingyu Lu, Zilin Jing, Fei Wang, Hoifung Poon, Tristan Naumann, Peniel Argaw·· 2 小时前AI 评分37
PICLE-PFN：首个支持动态治疗策略的纵向因果效应估计基础模型
A Foundation Model for Policy In-Context Longitudinal Causal Effect Estimation
AI 导读
研究者提出 PICLE-PFN，首个面向纵向因果效应估计、支持动态治疗策略的基础模型。该模型基于从纵向结构因果模型采样的合成数据预训练，通过上下文历史-决策示例表示目标策略，一次前向传播即可直接预测条件平均潜在结果（CAPOs），无需针对具体任务训练或调参。
正文 · 原文
Abstract:Longitudinal causal effect estimation is complicated by time-varying confounding and treatment-confounder feedback loops. Existing methods generally require data-specific nuisance-model fitting, and many are tailored to prespecified treatment sequences or prediction horizons, leading to time-consuming computation at deployment. We introduce *Policy In-Context Longitudinal causal Effect estimation with Prior-data Fitted Networks* (**PICLE-PFN**), the first foundation model for longitudinal causal effect estimation that supports *dynamic* treatment policies. PICLE-PFN represents target policies through in-context history-decision demonstrations, providing a unified interface for a broad class of treatment policies. Pretrained over synthetic data sampled from longitudinal structural causal models, PICLE-PFN amortizes causal effect estimation across datasets, policies, and horizons. Given a new observational dataset and target policy, it directly predicts conditional average potential outcomes (CAPOs) in one forward pass without task-specific training or hyperparameter tuning. Across synthetic and semi-synthetic longitudinal benchmarks, PICLE-PFN achieves competitive RMSE and substantially improved uncertainty calibration with a fraction of deployment runtime, compared to state-of-the-art conventional estimators trained and tuned separately for each dataset.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.32863 [cs.AI] |
| (or arXiv:2610.32863v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32863 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Wenxin Chen [view email]
[v1]
Fri, 23 Oct 2026 16:34:49 UTC (2,643 KB)
来源：arXiv cs.AI · arxiv.org