RankGR: Rank-Enhanced Generative Retrieval with Listwise Direct Preference Optimization in Recommendation DOI: XXXXXXX.XXXXXXX Conference: Make sure to enter the correct conference title from your rights confirmation email; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/2018/06 CCS: Informa
Kairui Fu, Changfa Wu, Kun Yuan, Binbin Cao, Dunxian Huang, Yuliang Yan, Junjun Zheng, Jianning Zhang, Silu Zhou, Jian Wu, Kun Kuang · 2026
Read RankGR: Rank-Enhanced Generative Retrieval with Listwise Direct Preference Optimization in Recommendation DOI: XXXXXXX.XXXXXXX Conference: Make sure to enter the correct conference title from your rights confirmation email; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/2018/06 CCS: Informa by Kairui Fu, Changfa Wu, Kun Yuan, Binbin Cao, Dunxian Huang, Yuliang Yan, Junjun Zheng, Jianning Zhang, Silu Zhou, Jian Wu, Kun Kuang (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-2602-08575.txt.
Description
Generative retrieval in recommenders usually decodes item identifiers token by token, treating each token as the sole target. Kairui Fu and ten co-authors argue that narrow focus misses the structure of user preferences and the interplay with behavior sequences. Their arXiv paper introduces RankGR, a rank-enhanced method built on listwise preference optimization.
Retrieval splits into two stages. An initial assessment phase learns hierarchical preferences and partial orders through listwise optimization. A refined scoring phase then re-ranks the top candidates against input sequences with a light scoring module. Both phases train jointly, and the deployed system handles nearly ten thousand requests per second. Offline and online tests, including Taobao's Guess You Like section, confirm the gains. Readers get a ranking-aware upgrade for generative recommendation.
Catalog record
| CREATOR | Kairui Fu, Changfa Wu, Kun Yuan, Binbin Cao, Dunxian Huang, Yuliang Yan, Junjun Zheng, Jianning Zhang, Silu Zhou, Jian Wu, Kun Kuang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-08575