# MARS: Modality-Aligned Retrieval for Sequence Augmented CTR Prediction

arXiv: https://arxiv.org/abs/2509.01184 · HTML: https://mlsanity.com/paper/2509.01184/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Sep 2025.

## Built on by (most cited first, 2 of 2)

- [ScaleToT: Generalizing Structured LLM Reasoning for Billion-Scale Low-Activity User Modeling](https://mlsanity.com/paper/2606.24605/) (2026)
- [Towards a Densing Law for User Representation Learning at Billion-Scale Capacity](https://mlsanity.com/paper/2608.23392/) (2026)

## Abstract (arXiv)

Click-through rate (CTR) prediction serves as a cornerstone of recommender systems. Despite the strong performance of current CTR models based on user behavior modeling, they are still severely limited by interaction sparsity, especially in low-active user scenarios. To address this issue, data augmentation of user behavior is a promising research direction. However, existing data augmentation methods heavily rely on collaborative signals while overlooking the rich multimodal features of items, leading to insufficient modeling of low-active users. To alleviate this problem, we propose a novel framework MARS (Modality-Aligned Retrieval for Sequence Augmented CTR Prediction). MARS utilizes a Stein kernel-based approach to align text and image features into a unified and unbiased semantic space to construct multimodal user embeddings. Subsequently, each low-active user's behavior sequence is augmented by retrieving, filtering, and concentrating the most similar behavior sequence of high-active users via multimodal user embeddings. Validated by extensive offline experiments and online A/B tests, our framework MARS consistently outperforms state-of-the-art baselines and achieves substantial growth on core business metrics within Kuaishou~footnotehttps://www.kuaishou.com/. Consequently, MARS has been successfully deployed, serving the main traffic for hundreds of millions of users. To ensure reproducibility, we provide anonymous access to the implementation code~footnotehttps://github.com/wangshukuan/MARS.