IDProxy: Cold-Start CTR Prediction for Ads and Recommendation at Xiaohongshu with Multimodal LLMs
This addresses the item cold-start issue for large-scale advertising and recommendation systems at Xiaohongshu, serving hundreds of millions of users daily, with an incremental improvement over existing ID-based methods.
The paper tackles the cold-start problem in CTR prediction for ads and recommendations by introducing IDProxy, which uses multimodal LLMs to generate proxy embeddings from content signals for new items, achieving effective CTR prediction without usage data and demonstrating success in offline experiments and online A/B tests at Xiaohongshu.
Click-through rate (CTR) models in advertising and recommendation systems rely heavily on item ID embeddings, which struggle in item cold-start settings. We present IDProxy, a solution that leverages multimodal large language models (MLLMs) to generate proxy embeddings from rich content signals, enabling effective CTR prediction for new items without usage data. These proxies are explicitly aligned with the existing ID embedding space and are optimized end-to-end under CTR objectives together with the ranking model, allowing seamless integration into existing large-scale ranking pipelines. Offline experiments and online A/B tests demonstrate the effectiveness of IDProxy, which has been successfully deployed in both Content Feed and Display Ads features of Xiaohongshu's Explore Feed, serving hundreds of millions of users daily.