GRAB: An LLM-Inspired Sequence-First Click-Through Rate Prediction Modeling Paradigm
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, Lin Liu · 2026
Read GRAB: An LLM-Inspired Sequence-First Click-Through Rate Prediction Modeling Paradigm by 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, Lin Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-01865.txt.
Description
Shaopeng Chen and colleagues build GRAB, a generative ad ranking system that predicts click through rates, fixing deep recommendation models' generalization and long sequence limits. Its Causal Action-aware Multi-channel Attention mechanism reads time patterns and action cues in user behavior. Full scale online deployment beats established deep recommendation models, with revenue up over three percent and click through rate up over three percent. Expressive power also grows monotonically and approximately linearly as longer interaction sequences are used.
Catalog record
| CREATOR | 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, Lin Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65645 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01865