Joint Model Parameter Scaling and Universal-Domain Data Integration for E-commerce Search Ranking
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914589452533760 |
|---|---|
| author | Yu, Liren Li, Caiyuan Dong, Feiyi Zhang, Tao Zhang, Zhixuan Ou, Dan Tang, Haihong Zheng, Bo |
| author_facet | Yu, Liren Li, Caiyuan Dong, Feiyi Zhang, Tao Zhang, Zhixuan Ou, Dan Tang, Haihong Zheng, Bo |
| contents | Scaling studies for industrial search, advertising, and recommendation have largely emphasized enlarging model capacity or refining architectures. Yet in real-world systems, performance is constrained not only by model size but also by the quality and distribution of training data. Our empirical analysis shows two key bottlenecks: increasing parameters alone yields progressively smaller gains, and the challenges introduced by heterogeneous, large-scale behavior data cannot be fully resolved by architecture tuning in isolation. To address this issue, we present UniScale, a unified framework that couples data scaling with model design. UniScale consists of two components. First, ES$^3$, an entire-space sample construction system, broadens supervision beyond conventional sampled training data by enriching intra-domain search contexts with globally attributed supervisory signals and introducing cross-domain examples that reflect user decisions under comparable content exposure conditions. Second, HHSFT, a heterogeneous hierarchical fusion transformer, is tailored to exploit the resulting large-scale heterogeneous data through hierarchical feature interaction and user-interest fusion across the entire behavior space. Together, these components enable more effective scaling than structure-centric optimization alone. Experiments show that UniScale consistently improves offline performance and demonstrates favorable scaling behavior. In online A/B tests on a large e-commerce search platform, it delivers a 1.70% increase in purchases and a 2.04% lift in GMV. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_24226 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Joint Model Parameter Scaling and Universal-Domain Data Integration for E-commerce Search Ranking Yu, Liren Li, Caiyuan Dong, Feiyi Zhang, Tao Zhang, Zhixuan Ou, Dan Tang, Haihong Zheng, Bo Information Retrieval Machine Learning Scaling studies for industrial search, advertising, and recommendation have largely emphasized enlarging model capacity or refining architectures. Yet in real-world systems, performance is constrained not only by model size but also by the quality and distribution of training data. Our empirical analysis shows two key bottlenecks: increasing parameters alone yields progressively smaller gains, and the challenges introduced by heterogeneous, large-scale behavior data cannot be fully resolved by architecture tuning in isolation. To address this issue, we present UniScale, a unified framework that couples data scaling with model design. UniScale consists of two components. First, ES$^3$, an entire-space sample construction system, broadens supervision beyond conventional sampled training data by enriching intra-domain search contexts with globally attributed supervisory signals and introducing cross-domain examples that reflect user decisions under comparable content exposure conditions. Second, HHSFT, a heterogeneous hierarchical fusion transformer, is tailored to exploit the resulting large-scale heterogeneous data through hierarchical feature interaction and user-interest fusion across the entire behavior space. Together, these components enable more effective scaling than structure-centric optimization alone. Experiments show that UniScale consistently improves offline performance and demonstrates favorable scaling behavior. In online A/B tests on a large e-commerce search platform, it delivers a 1.70% increase in purchases and a 2.04% lift in GMV. |
| title | Joint Model Parameter Scaling and Universal-Domain Data Integration for E-commerce Search Ranking |
| topic | Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2603.24226 |