解决推荐系统中停留时间预测被点击率误导的问题。
ORCA: Mitigating Over-Reliance for Multi-Task Dwell Time Prediction with Causal Decoupling
- 用因果解耦方法分离点击率与停留时间的虚假关联
- 提升中等停留时长预测准确率，平均提升10.6%
- 不依赖特定模型，适合实际推荐系统部署
停留时间（DT）是评估推荐系统用户偏好的关键后点击指标，可补充传统点击率（CTR）。尽管多任务学习被广泛用于联合优化DT与CTR，但我们发现多任务模型会系统性地将DT预测集中在最短和最长区间，低估中等时长。这源于对CTR-DT虚假相关性的过度依赖。为此，我们提出ORCA，通过因果解耦缓解该问题。具体地，ORCA显式建模并消除CTR带来的负面迁移，同时保留其正面迁移。进一步引入（i）特征级反事实干预，和（ii）带实例逆权重的任务交互模块，削弱CTR的中介效应，恢复停留时间的直接语义。ORCA具有模型无关性，易于部署。实验显示，DT指标平均提升10.6%，且不影响CTR。代码已开源。
原文摘要 · Abstract (English)
Dwell time (DT) is a critical post-click metric for evaluating user preference in recommender systems, complementing the traditional click-through rate (CTR). Although multi-task learning is widely adopted to jointly optimize DT and CTR, we observe that multi-task models systematically collapse their DT predictions to the shortest and longest bins, under-predicting the moderate durations. We attribute this moderate-duration bin under-representation to over-reliance on the CTR-DT spurious correlation, and propose ORCA to address it with causal-decoupling. Specifically, ORCA explicitly models and subtracts CTR's negative transfer while preserving its positive transfer. We further introduce (i) feature-level counterfactual intervention, and (ii) a task-interaction module with instance inverse-weighting, weakening CTR-mediated effect and restoring direct DT semantics. ORCA is model-agnostic and easy to deploy. Experiments show an average 10.6% lift in DT metrics without harming CTR. Code is available at https://github.com/Chrissie-Law/ORCA-Mitigating-Over-Reliance-for-Multi-Task-Dwell-Time-Prediction-with-Causal-Decoupling.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。