arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.01865cs.IRcs.AI
收藏

GRAB: 一种受大语言模型启发的序列优先点击通过率预测建模范式

GRAB: An LLM-Inspired Sequence-First Click-Through Rate Prediction Modeling Paradigm

Baidu Inc.(百度公司)

Shaopeng Chen, Chuyue Xie, Huimin Ren, Shaozong Zhang, Han Zhang, Ruobing Cheng, Zhiqiang Cao, Zehao Ju, Yu Gao, Jie Ding, Xiaodong Chen, Xuewu Jiao, Shuanglong Li, Liu Lin

AI总结:

GRAB通过引入因果动作感知多通道注意力机制,实现了在点击通过率预测上的性能提升,并在实际部署中实现了更高的收入和CTR。

摘要:

传统深度学习推荐模型(DLRMs)在性能和效率方面面临日益严峻的瓶颈,常常难以应对泛化能力和长序列建模问题。受大语言模型(LLMs)规模化成功的启发,我们提出了百度广告生成排序(GRAB),一种端到端生成框架用于点击通过率(CTR)预测。GRAB集成了一个新颖的因果动作感知多通道注意力(CamA)机制,以有效捕捉用户行为序列中的时间动态和特定动作信号。全规模在线部署显示,GRAB显著优于现有DLRMs,实现了3.05%的收入增长和3.49%的CTR提升。此外,该模型表现出良好的扩展性:其表达能力随着使用更长的交互序列而呈现单调且近似线性的提升。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑