| title | OneTrans-GSR: A Generative Slate Ranking Paradigm Unifying Reranking and Mix-Ranking with One Autoregressive Transformer |
|---|---|
| authors | Feng Cheng, Yu-Hsuan Huang, Fengyi Wang, Chaofan Zhou, Zhennan Zhu, Huan He, Bo Zhang, Ruoqi Zhang, Zhanfeng Mo, Ziwei Liu, Lanyi Li, Yifeng Pan, Junhua Long, Ziqiu Zhou, Yue Lin, Shaowei Liu |
| date | 2026-11-03 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2611.04720 |
| affiliation | ByteDance |
| upvotes | 36 (Hugging Face Daily Papers, 2026-11-11) |
| links | abs pdf hf |
| cite | |
| related | DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation Dockerless: Environment-Free Program Verifier for Coding Agents SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces SwanTale: Unified Multi-Speaker Speech and Audio Generation for Instruct and Zero-Shot Tasks |
| abstract | Industrial recommender systems allocate substantial computation to pointwise candidate scoring, yet separate reranking and mix-ranking stages still assemble the final exposure slate using hand-crafted rules and specialized low-capacity models. Mix-ranking combines product cards, short videos, livestreams, and advertising cards supplied by distinct upstream business channels. Although these rules and models encode valuable domain knowledge, they are inherently limited in leveraging additional data and computation. They constrain the combinatorial space of candidate slates and determine objective trade-offs through manual tuning. To enable scalable and effective slate generation, we reformulate reranking and mix-ranking as a unified learning problem whose performance can improve with computation. Specifically, we instantiate this formulation across the reranking and mix-ranking stages and propose OneTrans-GSR, a single decoder-only policy that extends our deployed OneTrans backbone to generate the exposure slate end-to-end. During training, a reward model evaluates complete slates and guides reinforcement learning toward slate-level and long-term objectives. During decoding, business constraints are explicitly enforced. Offline evaluation shows that slate quality consistently improves as reward-model and policy compute increase. Deployed in a large-scale industrial recommendation system, OneTrans-GSR mixes heterogeneous content types within a unified slate. In online A/B tests, it improves GMV/u by 2.63%, advertiser value by 2.06%, and repurchase rate by 2.03%, while reducing end-to-end latency across the reranking and mix-ranking stages by 13%. These results mark a paradigm shift in slate generation: moving beyond human-engineered heuristics toward a general learned policy creates a scalable path for further algorithmic optimization and compute-driven improvement. |
bibtex
@misc{cheng2026onetrans,
title={OneTrans-GSR: A Generative Slate Ranking Paradigm Unifying Reranking and Mix-Ranking with One Autoregressive Transformer},
author={Feng Cheng and Yu-Hsuan Huang and Fengyi Wang and Chaofan Zhou and Zhennan Zhu and Huan He and Bo Zhang and Ruoqi Zhang and Zhanfeng Mo and Ziwei Liu and Lanyi Li and Yifeng Pan and Junhua Long and Ziqiu Zhou and Yue Lin and Shaowei Liu},
year={2026},
eprint={2611.04720},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2611.04720}
}