Metraipolis › archive › dc_7e80a524e648

ReuseGR: Offline--Online Decoupled Generative Retrieval via Adaptive Prefix Reuse — arXiv:2611.04259

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04259
License (as archived)
CC0-1.0
Archived
2026-11-04T19:02:49.545Z
SHA-256 of the archived text
947828406cc35f0405f3b20ee657eb7ae54cc0ca04cab39dd46c84e20ae4f0bc
Stable URL of this page
https://metraipolis.com/documento/dc_7e80a524e648

How to cite

ReuseGR: Offline--Online Decoupled Generative Retrieval via Adaptive Prefix Reuse — arXiv:2611.04259 — arXiv (en), https://arxiv.org/abs/2611.04259 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7e80a524e648 (Metraipolis archive dc_7e80a524e648, sha256 947828406cc35f0405f3b20ee657eb7ae54cc0ca04cab39dd46c84e20ae4f0bc)
BibTeX
@misc{dc7e80a524e648,
  title = {ReuseGR: Offline--Online Decoupled Generative Retrieval via Adaptive Prefix Reuse — arXiv:2611.04259},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04259},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7e80a524e648; sha256 947828406cc35f0405f3b20ee657eb7ae54cc0ca04cab39dd46c84e20ae4f0bc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04259. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Larger generative retrieval models can improve retrieval quality, but their decoding cost makes low-latency serving difficult. When different users submit the same query, generating complete Semantic IDs (SIDs) for every request repeats computation, while their interaction histories call for different results. The challenge is to reuse large-model computation across requests while keeping retrieval personalized. We propose \textbf{ReuseGR}, which separates query-level candidate planning from request-level personalized completion. A 7B offline planner generates SID prefix sets shared across requests with the same query. A 0.5B online model uses each request's history to select prefixes and generate the remaining tokens. ReuseCut chooses a cache depth for each query based on historical coverage, balancing computation reuse and candidate coverage. To strengthen online completion, we transfer 7B suffix-scoring knowledge to 0.5B through path-level knowledge distillation (Path KD). BudgetCut then selects scored branches to improve final top-100 recall under a fixed branch budget. Its selection policy is initialized by on-policy distillation (OPD) and refined through group-relative policy optimization (GRPO). On an industrial search-advertising dataset with approximately 2M query--item records and 3.38M catalog items, ReuseGR improves query-level HR@100 by 3.83 percentage points and reduces mean online latency by 38.9\% relative to an all-online 0.5B baseline when prefixes are prepared. It achieves HR@100 comparable to an all-online 7B baseline with 57.7\% lower online latency. A 14-day user-level online A/B test against the 0.5B baseline shows relative gains of 0.93\% in click-through rate and 1.05\% in revenue per thousand requests.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.IR, nuovi lavori del giorno).
Categorie arXiv: Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2611.04259 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-04T18:39:23.549Z dalla pagina https://arxiv.org/list/cs.IR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).