Metraipolis › archive › dc_ca9cb69a9768

Predicting What Answer Engines Cite: Open Decision Models, Rerankers and a Distilled 4B Student on Ten Engines under Controlled Retrieval — arXiv:2610.38090

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.38090
License (as archived)
CC0-1.0
Archived
2026-10-31T06:11:18.841Z
SHA-256 of the archived text
da7df0a9b9ccf20dbe06f1bf91da6d4378b0c9135aa4ee7a447920c480640d3f
Stable URL of this page
https://metraipolis.com/documento/dc_ca9cb69a9768

How to cite

Predicting What Answer Engines Cite: Open Decision Models, Rerankers and a Distilled 4B Student on Ten Engines under Controlled Retrieval — arXiv:2610.38090 — arXiv (en), https://arxiv.org/abs/2610.38090 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ca9cb69a9768 (Metraipolis archive dc_ca9cb69a9768, sha256 da7df0a9b9ccf20dbe06f1bf91da6d4378b0c9135aa4ee7a447920c480640d3f)
BibTeX
@misc{dcca9cb69a9768,
  title = {Predicting What Answer Engines Cite: Open Decision Models, Rerankers and a Distilled 4B Student on Ten Engines under Controlled Retrieval — arXiv:2610.38090},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.38090},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ca9cb69a9768; sha256 da7df0a9b9ccf20dbe06f1bf91da6d4378b0c9135aa4ee7a447920c480640d3f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.38090. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Generative engine optimization needs a scorer that ranks candidate pages by the share of an answer they will receive as citations. We measure pretrained rerankers, a hosted typed decision model (Jev), eight open decision models and distilled students on GEO-Bench queries answered by ten engines that receive the same candidate sources in their prompt. Where several engines answer the same query, they agree with one another at a within-group Spearman correlation of 0.704. Jev and two open 27B decision models, used without training, are the strongest scorers at 0.575 to 0.578. A Qwen3-Reranker-4B fine-tuned with low-rank adapters on teacher labels for other GEO-Bench queries, released as GEO Citation Ranker 4B, reaches 0.542 to 0.549 across three seeds, and its differences with the 27B models and with Jev are not distinguishable from zero on 229 queries. Most of the remaining distance between the student and Jev disappears when Jev scores one passage at a time, as the student does. Students trained on teacher labels and on observed engine citations reach the same level, and labels from Jev alone match the two-teacher target at 0.6B. We release the model, the benchmark, the predictions of every reported scorer and the code.

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).
Pagina del lavoro: https://arxiv.org/abs/2610.38090 — 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 2026-10-31T06:07:08.624Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).