Traditional Deep Learning Recommendation Models (DLRMs) face increasing bottlenecks in performance and efficiency, often struggling with generalization and long-sequence modeling. Inspired by the scaling success of Large Language Models (LLMs), we propose Generative Ranking for Ads at Baidu (GRAB), an end-to-end generative framework for Click-Through Rate (CTR) prediction. GRAB integrates a novel Causal Action-aware Multi-channel Attention (CamA) mechanism to effectively capture temporal dynamics and specific action signals within user behavior sequences. Full-scale online deployment demonstrates that GRAB significantly outperforms established DLRMs, delivering a 3.05% increase in revenue and a 3.49% rise in CTR. Furthermore, the model demonstrates desirable scaling behavior: its expressive power shows a monotonic and approximately linear improvement as longer interaction sequences are utilized.
@article{arxiv.2602.01865,
title = {GRAB: An LLM-Inspired Sequence-First Click-Through Rate Prediction Modeling Paradigm},
author = {Shaopeng Chen and Chuyue Xie and Huimin Ren and Shaozong Zhang and Han Zhang and Ruobing Cheng and Zhiqiang Cao and Zehao Ju and Yu Gao and Jie Ding and Xiaodong Chen and Xuewu Jiao and Shuanglong Li and Liu Lin},
journal= {arXiv preprint arXiv:2602.01865},
year = {2026}
}