# Towards Efficient Whole Page Ranking with One-Pass Inference

**Authors:** Qi Xu, Zhigang Hua, Zijie Lei, Yan Xie, Jizhe Zhang, Lynn Lu, Delun Du, Sha Hua, Ren Mao, Ke Pan, Haomin Yu, Sandeep Pandey, Luke Simon  
**Lab:** Meta · **Published:** 3 Nov 2026  
**Links:** [arXiv](https://arxiv.org/abs/2611.05542) · [PDF](https://arxiv.org/pdf/2611.05542) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2611.05542.html)

## Key numbers

- **4.07×**: …the generation quality of an autoregressive CGA baseline while achieving 4.07x and 3.54x faster CPU decoding
- **0.104%**: MIS improves the realization metric by 0.104% and watch time by 0.058%
- **0.058%**: MIS improves the realization metric by 0.104% and watch time by 0.058%

## Abstract

Whole-page ranking jointly optimizes advertisements and organic content to balance business outcomes and user engagement under complex layout and business constraints. Existing non-autoregressive (NAR) approaches enable efficient page generation but typically score candidate items independently of previously selected items, limiting their ability to capture order-dependent interactions. Autoregressive (AR) methods address this limitation through sequential conditioning, but incur substantial inference overhead that challenges large-scale deployment. Bridging this quality-efficiency gap remains a challenge in industrial whole-page ranking. We propose Multi-Item Scoring (MIS), a prefix-aware NAR-style framework that incorporates sequential context into efficient multi-item scoring. MIS integrates global candidate interactions, localized prefix conditioning, and position-aware scoring within a shared representation architecture. By computing reusable representations once per request and performing lightweight prefix-conditioned scoring during constrained beam search, MIS captures important sequential dependencies without repeated full-model inference. Experiments on KuaiRand-1K and RL4RS demonstrate that MIS approaches the generation quality of an autoregressive CGA baseline while achieving 4.07x and 3.54x faster CPU decoding, respectively. In a randomized evaluation on a large-scale social-video platform, MIS improves the realization metric by 0.104% and watch time by 0.058%, while reducing ranking-service CPU consumption by 9.73%. These results demonstrate that prefix-aware multi-item scoring can narrow the quality gap between NAR and AR generation while retaining the efficiency required for industrial scalability.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.