{
  "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.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2611.05542",
  "author_total": 13,
  "authors": [
    {
      "name": "Qi Xu"
    },
    {
      "name": "Zhigang Hua"
    },
    {
      "name": "Zijie Lei"
    },
    {
      "name": "Yan Xie"
    },
    {
      "name": "Jizhe Zhang"
    },
    {
      "name": "Lynn Lu"
    },
    {
      "name": "Delun Du"
    },
    {
      "name": "Sha Hua"
    },
    {
      "name": "Ren Mao"
    },
    {
      "name": "Ke Pan"
    },
    {
      "name": "Haomin Yu"
    },
    {
      "name": "Sandeep Pandey"
    },
    {
      "name": "Luke Simon"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/ap6gsc3ebwwzd.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2611.05542",
  "fulltext_url": null,
  "id": "ap6gsc3ebwwzd",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2611.05542",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.AI"
    ],
    [
      "arxiv",
      "stat.ML"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://arxiv.org/pdf/2611.05542",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-03",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/ap6gsc3ebwwzd.ris",
  "sources": [
    {
      "first_seen": "2026-11-05T15:21:06Z",
      "native_id": "2611.05542",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2611.05542"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Towards Efficient Whole Page Ranking with One-Pass Inference",
  "title_text": "Towards Efficient Whole Page Ranking with One-Pass Inference",
  "type": "preprint",
  "updated_at": "2026-11-09T00:20:32Z",
  "url": "https://research.draconitic.com/paper/ap6gsc3ebwwzd",
  "venue": "arXiv",
  "versions": []
}