跳到正文
arXiv cs.AI· Bojing Feng, Jinqiu Deng, Zihan Hong, Yuan Xue, Wenqi Xiang, Shuhan Wang, Feng Xiao, Shiwei Wang, Zijiang Liu, Shaomin Li, Yongzhi Qi·· 4 小时前AI 评分29

LULR:为基于 LLM 的增益估计正则化处理泄漏

Regularizing Treatment Leakage for LLM-based Uplift Estimation

AI 导读

研究者提出 LULR(LLM-based Uplift Leakage Regularization),在冻结 LLM 编码器的前提下,用 CLUB 互信息上界的变分近似惩罚因果头共享表示与处理变量的依赖,而非笼统的参数收缩。

正文 · 原文

Authors:Bojing Feng, Jinqiu Deng, Zihan Hong, Yuan Xue, Wenqi Xiang, Shuhan Wang, Feng Xiao, Shiwei Wang, Zijiang Liu, Shaomin Li, Yongzhi Qi

View PDF HTML (experimental)

Abstract:Large language models (LLMs) provide semantic features for uplift estimation, but treatment-revealing text can induce prediction shortcuts that do not transfer to deployment. We propose \emph{LLM-based Uplift Leakage Regularization} (\textsc{LULR}) to address this risk. With the LLM encoder frozen, \textsc{LULR} penalizes dependence between the causal head's trainable shared representation and treatment while retaining separate outcome heads. The penalty uses a variational approximation to the contrastive log-ratio upper bound (CLUB) of mutual information, targeting treatment dependence rather than generic parameter shrinkage. Controlled proxy experiments provide exploratory evidence of improved ranking scores under injected treatment information relative to no regularization and the tested weight-decay setting. Structured-only benchmarks show modest, dataset-dependent effects. On an industrial pricing randomized controlled trial, mean offline $\text{lift@}0.3$ is $0.258$ for \textsc{LULR}, compared with $0.106$ for the same-backbone baseline. A separate one-week online A/B test of the complete deployed pricing policy records $5.45\%$ higher sales volume and $1.91\%$ higher gross merchandise value than the control arm. These short-term policy-level gains do not isolate the regularizer's effect.
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.20794 [cs.AI]
  (or arXiv:2611.20794v2 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.20794

arXiv-issued DOI via DataCite

Submission history

From: Bojing Feng [view email]
[v1] Mon, 16 Nov 2026 09:42:24 UTC (86 KB)
[v2] Tue, 17 Nov 2026 06:13:58 UTC (86 KB)

来源:arXiv cs.AI · arxiv.org