UniScale: Synergistic Entire Space Data and Model Scaling for Search Ranking

2026-03-25

AI Summary

This work addresses the diminishing returns of simply scaling up model parameters in search ranking tasks and the challenge of effectively modeling complex, heterogeneous data distributions through architectural adjustments alone. To overcome these limitations, the authors propose UniScale, a novel framework that enables the first coordinated scaling of both data and model architecture. UniScale integrates a high-quality, large-scale training dataset constructed via the E-commerce-wide Sample System (ES³) with a Heterogeneous Hierarchical Sample Fusion Transformer (HHSFT) designed to accurately capture intricate user interests and data heterogeneity. Extensive experiments on a large-scale e-commerce search platform demonstrate that the proposed approach significantly improves key business metrics and exhibits clear synergistic gains from the joint scaling of data and model capacity.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Scalability of ML Systems
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Recent advances in Large Language Models (LLMs) have inspired a surge of scaling law research in industrial search, advertising, and recommendation systems. However, existing approaches focus mainly on architectural improvements, overlooking the critical synergy between data and architecture design. We observe that scaling model parameters alone exhibits diminishing returns, i.e., the marginal gain in performance steadily declines as model size increases, and that the performance degradation caused by complex heterogeneous data distributions is often irrecoverable through model design alone. In this paper, we propose UniScale to address these limitation, a novel co-design framework that jointly optimizes data and architecture to unlock the full potential of model scaling, which includes two core parts: (1) ES$^3$ (Entire-Space Sample System), a high-quality data scaling system that expands the training signal beyond conventional sampling strategies from both intra-domain request contexts with global supervised signal constructed by hierarchical label attribution and cross-domain samples aligning with the essence of user decision under similar content exposure environment in search domain; and (2) HHSFT (Heterogeneous Hierarchical Sample Fusion Transformer), a novel architecture designed to effectively model the complex heterogeneous distribution of scaled data and to harness the entire space user behavior data with Heterogeneous Hierarchical Feature Interaction and Entire Space User Interest Fusion, thereby surpassing the performance ceiling of structure-only model tuning. Extensive experiments on large-scale real world E-commerce search platform demonstrate that UniScale achieves significant improvements through the synergistic co-design of data and architecture and exhibits clear scaling trends, delivering substantial gains in key business metrics.

Problem

• search ranking
• scaling laws
• data-architecture co-design
• heterogeneous data distribution
• model scaling

Innovation

• data-model co-design
• entire-space sampling
• heterogeneous hierarchical modeling
• scaling laws
• search ranking