Summary
Figuring out the best way to offer coupons for many types of services at once is tricky because discounts affect customer choices in complex ways. The authors developed TaQFormer, a type of AI model that treats each discount category as a special input, allowing it to understand how different discounts work together. They first train the model on large logs of past behavior and then adjust it with precise experiments to make better predictions that boost business value. Their approach led to a small but meaningful increase in user value without higher subsidy costs and was successfully used in real-world online marketplace coupon allocation.
What this means in practice
- •For e-commerce platform operators: Optimize multi-category discount strategies using multidimensional treatment models to increase user value without raising subsidy costs.
- •For online advertising teams: Integrate observational and experimental data to improve cross-category campaign targeting based on user response patterns.
Authors
Jiahong Li, Zeqin Yang, Yu Xiang, Zhifan Jiang, Jingqi Ma, Enzheng Hua, Jixing Xu, Jun Wu, Fei Wang, Enyun Yu, Jiecheng Guo
Abstract
We study coupon allocation on an online marketplace offering multiple service categories. Issuing coupons requires choosing a vector of category-specific discounts, accounting for both their direct effects and cross-category demand shifts. We present TaQFormer, a treatment-as-query Transformer for this multidimensional treatment setting. Each treatment coordinate is represented as a query token, allowing cross-attention to estimate the response to the full treatment vector. To combine large but policy-biased observational logs with a smaller stream of randomized controlled trial data, we use a two-stage pipeline: observational pretraining learns general outcome patterns, followed by LoRA fine-tuning on RCT data with a request-specific fusion gate. We evaluate the resulting system through the checks required for production, including offline uplift metrics, treatment-parallel serving, latency monitoring, and a 14-day online A/B test. At essentially the same subsidy cost rate, \model increases business value per user by 1.26\% over the production baseline and significantly outperforms both the production baseline and the RCT-only model, while meeting the latency budget. It was subsequently rolled out to all marketplace traffic and has operated at full scale for several months as of October 2026. The deployment illustrates how multidimensional treatment modeling and observational pretraining, combined with adaptation using RCT data, can support effective joint discount decisions in production.