发表机构
Southern University of Science and Technology; University of Pennsylvania（南方科技大学; 宾夕法尼亚大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对分布外推荐的风险问题，提出GenCAR方法，通过控制代理标签错误发现率，提升了不同基准下的分布外推荐候选恢复效果。
AI 中文摘要
在分布偏移场景下提供有用推荐，对于平衡分布外（OOD）推荐的效用与风险至关重要。然而，多数现有OOD方法仅优化排序或构建反事实候选，未控制服务集合的代理标签错误发现率（FDR）。本研究将OOD服务问题形式化为α-有效反事实推荐（α-VCR）问题，以保留从反事实监督中学习到的候选支持，同时控制代理标签FDR，并提出GenCAR方法，该方法结合基于偏好的反事实监督与校准集合选择。具体而言，GenCAR固定稳定偏好表示，同时对环境因素进行干预，通过偏好锚点和信任半径过滤将离线大语言模型的提议落地，并使用共形p值进行Benjamini–Hochberg选择。我们从理论上对条件反事实近似误差进行了界定，证明在可交换性和正回归依赖下，有限样本、无分布的代理标签FDR控制，以及在任意依赖下的Benjamini–Yekutieli保证。大量实验验证了实际代理错误发现比例，结果表明GenCAR在不同基准上始终提升OOD候选恢复效果。
英文摘要
Serving useful recommendations under distribution shift is crucial for balancing utility and risk in out-of-distribution (OOD) recommendation. However, most existing OOD methods improve ranking or construct counterfactual candidates without controlling the proxy-label false discovery rate (FDR) of the served set. In this work, we formulate OOD serving as the $α$-Valid Counterfactual Recommendation ($α$-VCR) problem to retain candidate support learned from counterfactual supervision while controlling proxy-label FDR, and propose GenCAR, which couples preference-grounded counterfactual supervision with calibrated set selection. In particular, GenCAR fixes the stable-preference representation while intervening on the environmental factor, grounds offline large language model proposals through preference anchors and trust-radius filtering, and uses conformal $p$-values for Benjamini--Hochberg selection. We theoretically bound conditional counterfactual approximation error and prove finite-sample, distribution-free control of proxy-label FDR under exchangeability and positive regression dependence, with a Benjamini--Yekutieli guarantee under arbitrary dependence. Extensive experiments audit realized proxy false discovery proportions and demonstrate that GenCAR consistently enhances OOD candidate recovery across diverse benchmarks.
Comments19 pages, 8 figures, 7 tables
