Metraipolis › archive › dc_db191b67344a
TS-GEAR: Taobao Shangou GEnerative Ad Retrieval — arXiv:2611.12180
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.12180
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-10T18:43:58.665Z
- SHA-256 of the archived text
bee8175cd2b27789668f4dd71058db3c1bfc158c30bdd11ab06ca187d2ec3825- Stable URL of this page
https://metraipolis.com/documento/dc_db191b67344a
How to cite
TS-GEAR: Taobao Shangou GEnerative Ad Retrieval — arXiv:2611.12180 — arXiv (en), https://arxiv.org/abs/2611.12180 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_db191b67344a (Metraipolis archive dc_db191b67344a, sha256 bee8175cd2b27789668f4dd71058db3c1bfc158c30bdd11ab06ca187d2ec3825)
BibTeX
@misc{dcdb191b67344a,
title = {TS-GEAR: Taobao Shangou GEnerative Ad Retrieval — arXiv:2611.12180},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.12180},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_db191b67344a; sha256 bee8175cd2b27789668f4dd71058db3c1bfc158c30bdd11ab06ca187d2ec3825},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.12180. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
On-demand delivery platforms such as DoorDash and Uber Eats impose a constraint absent in traditional e-commerce: merchants beyond the user's delivery range are invalid candidates rather than merely low-ranked. Generative retrieval has recently shown strong potential in large-scale retrieval systems. Adapting it to this setting, however, poses three challenges. First, existing semantic IDs encode geography only as a soft signal, so reachability can only rely on a non-differentiable post-generation hard truncation. Second, user preferences vary substantially across both regions and queries, so aggregating historical behaviors without conditioning on geography and the current query yields biased interest estimates. Third, the retrieved candidate set is evaluated by its collective value, yet autoregressive training scores each sequence independently, with no set-level signal to discourage redundant near-duplicates. To address the challenges, we propose TS-GEAR, a unified generative retrieval framework for on-demand delivery: (1) Geo-Semantic ID splits merchant identifiers into geographic prefixes and semantic suffixes, making reachability a prediction target rather than a post-hoc filter; (2) Query-Geo Target Attention aggregates historical behaviors using the real-time query and location as dual targets, while encoding each behavior's own location into its attention key; (3) GMPO (Group Marginal Policy Optimization) rewrites advantages using a deduplicated group reward and marginal advantage within GRPO, avoiding auxiliary diversity objectives. Deployed online with a lightweight 0.1B backbone, TS-GEAR serves over 150 million queries per day in the search advertising pipeline of Taobao Shangou, consistently outperforming state-of-the-art generative retrieval methods on both HitRate and geographic validity, and achieving 3.8% gains in ad impressions and 4.2% in ad revenue. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Information Retrieval (cs.IR) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2611.12180 — 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 2026-11-10T18:37:37.683Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).