GIST: Cross-Domain Click-Through Rate Prediction via Guided Content-Behavior Distillation

2025-07-07

AI Summary

To address data sparsity and cold-start challenges in cross-domain click-through rate (CTR) prediction for online advertising—particularly under real-world industrial conditions involving distribution shift and continual arrival of new data—this paper proposes GIST. Methodologically, GIST (1) decouples source- and target-domain training and introduces a Content-Behavior Joint Training (CBJT) module to align multimodal distributions; (2) designs an Asymmetric Similarity Integration (ASI) strategy enhanced by guided information to ensure stable representation learning and continual adaptation; and (3) establishes a cross-domain sequential knowledge distillation framework. Extensive offline evaluations and online A/B tests on the Xiaohongshu platform demonstrate that GIST significantly outperforms existing state-of-the-art methods. Deployed in a production advertising system serving hundreds of millions of users daily, GIST improves both CTR prediction accuracy and key business metrics.

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: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Cross-domain Click-Through Rate prediction aims to tackle the data sparsity and the cold start problems in online advertising systems by transferring knowledge from source domains to a target domain. Most existing methods rely on overlapping users to facilitate this transfer, often focusing on joint training or pre-training with fine-tuning approach to connect the source and target domains. However, in real-world industrial settings, joint training struggles to learn optimal representations with different distributions, and pre-training with fine-tuning is not well-suited for continuously integrating new data. To address these issues, we propose GIST, a cross-domain lifelong sequence model that decouples the training processes of the source and target domains. Unlike previous methods that search lifelong sequences in the source domains using only content or behavior signals or their simple combinations, we innovatively introduce a Content-Behavior Joint Training Module (CBJT), which aligns content-behavior distributions and combines them with guided information to facilitate a more stable representation. Furthermore, we develop an Asymmetric Similarity Integration strategy (ASI) to augment knowledge transfer through similarity computation. Extensive experiments demonstrate the effectiveness of GIST, surpassing SOTA methods on offline evaluations and an online A/B test. Deployed on the Xiaohongshu (RedNote) platform, GIST effectively enhances online ads system performance at scale, serving hundreds of millions of daily active users.

Problem

• Address data sparsity and cold start in cross-domain CTR prediction
• Improve knowledge transfer without overlapping user dependency
• Enhance representation stability via content-behavior alignment

Innovation

• Decouples source and target domain training
• Introduces Content-Behavior Joint Training Module
• Uses Asymmetric Similarity Integration strategy