Generate the browsing process for short-video recommendation
This work addresses the problem of predicting user engagement for short video platforms like Kuaishou, offering an incremental improvement over existing methods.
The paper tackles watch time prediction in short video recommendation by proposing a generative method that simulates users' watching journey using collaborative information and interaction behaviors, achieving state-of-the-art performance with a +0.13% improvement in APP duration and 83% XAUC on industrial datasets.
This paper proposes a generative method to dynamically simulate users' short video watching journey for watch time prediction in short video recommendation. Unlike existing methods that rely on multimodal features for video content understanding, our method simulates users' sustained interest in watching short videos by learning collaborative information, using interest changes from existing positive and negative feedback videos and user interaction behaviors to implicitly model users' video watching journey. By segmenting videos based on duration and adopting a Transformer-like architecture, our method can capture sequential dependencies between segments while mitigating duration bias. Extensive experiments on industrial-scale and public datasets demonstrate that our method achieves state-of-the-art performance on watch time prediction tasks. The method has been deployed on Kuaishou Lite, achieving a significant improvement of +0.13\% in APP duration, and reaching an XAUC of 83\% for single video watch time prediction on industrial-scale streaming training sets, far exceeding other methods. The proposed method provides a scalable and effective solution for video recommendation through segment-level modeling and user engagement feedback.