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

TL;DR

GIST is a novel cross-domain CTR prediction model that decouples training processes, aligns content-behavior distributions, and employs similarity strategies to improve knowledge transfer, outperforming existing methods in both offline and online settings.

Contribution

The paper introduces GIST, a lifelong sequence model with Content-Behavior Joint Training and Asymmetric Similarity Integration, addressing limitations of joint training and pre-training in cross-domain CTR prediction.

Findings

GIST outperforms state-of-the-art methods in offline evaluations.

GIST significantly improves online ad system performance.

Deployment on Xiaohongshu demonstrates large-scale effectiveness.

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.