Adaptive Domain Scaling for Personalized Sequential Modeling in Recommenders

2025-02-08

AI Summary

In super-apps (e.g., TikTok), sequence recommendation suffers from modeling bias due to large inter-domain behavioral discrepancies and complex, heterogeneous user intents across diverse business scenarios. Method: This paper proposes Adaptive Domain Scaling (ADS), a novel framework that jointly models user behavior sequences and candidate items from a multi-domain collaborative perspective. ADS introduces domain-aware attention and a learnable scaling mechanism to enable target-aware, dynamic cross-domain intent understanding. Its architecture comprises two core modules—Personalized Sequence Representation Generation (PSRG) and Personalized Candidate Representation Generation (PCRG)—designed for end-to-end training. Contribution/Results: ADS achieves significant improvements in Recall@10 and AUC on both public benchmarks and industrial-scale datasets (tens of billions of samples). Deployed online in TikTok’s advertising and e-commerce scenarios, it yields +8.2% CTR and +5.7% GMV, demonstrating strong effectiveness and practical applicability.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Application Domains: Humanities & Computational Social Science
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems

Abstract

Users generally exhibit complex behavioral patterns and diverse intentions in multiple business scenarios of super applications like Douyin, presenting great challenges to current industrial multi-domain recommenders. To mitigate the discrepancies across diverse domains, researches and industrial practices generally emphasize sophisticated network structures to accomodate diverse data distributions, while neglecting the inherent understanding of user behavioral sequence from the multi-domain perspective. In this paper, we present Adaptive Domain Scaling (ADS) model, which comprehensively enhances the personalization capability in target-aware sequence modeling across multiple domains. Specifically, ADS comprises of two major modules, including personalized sequence representation generation (PSRG) and personalized candidate representation generation (PCRG). The modules contribute to the tailored multi-domain learning by dynamically learning both the user behavioral sequence item representation and the candidate target item representation under different domains, facilitating adaptive user intention understanding. Experiments are performed on both a public dataset and two billion-scaled industrial datasets, and the extensive results verify the high effectiveness and compatibility of ADS. Besides, we conduct online experiments on two influential business scenarios including Douyin Advertisement Platform and Douyin E-commerce Service Platform, both of which show substantial business improvements. Currently, ADS has been fully deployed in many recommendation services at ByteDance, serving billions of users.

Problem

• Enhance multi-domain recommendation personalization
• Understand user behavior across diverse domains
• Adaptive modeling for complex user intentions

Innovation

• Adaptive Domain Scaling (ADS) model
• Personalized sequence representation generation
• Personalized candidate representation generation