Metraipolis › archive › dc_cf7decb2120b
GroundedGEO: Auditing the Evidence Gap in Generative Search Rankings — arXiv:2609.25189
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.25189
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T13:46:13.949Z
- SHA-256 of the archived text
aab5d3073aec7e85aa4035d1aa09559cfcb60597d202c6055823e6ba3fe767fa- Stable URL of this page
https://metraipolis.com/documento/dc_cf7decb2120b
How to cite
GroundedGEO: Auditing the Evidence Gap in Generative Search Rankings — arXiv:2609.25189 — arXiv (en), https://arxiv.org/abs/2609.25189 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cf7decb2120b (Metraipolis archive dc_cf7decb2120b, sha256 aab5d3073aec7e85aa4035d1aa09559cfcb60597d202c6055823e6ba3fe767fa)
BibTeX
@misc{dccf7decb2120b,
title = {GroundedGEO: Auditing the Evidence Gap in Generative Search Rankings — arXiv:2609.25189},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.25189},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cf7decb2120b; sha256 aab5d3073aec7e85aa4035d1aa09559cfcb60597d202c6055823e6ba3fe767fa},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25189. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Generative search systems rank products and services for consequential decisions, and publishers can cheaply make candidate text look relevant. Yet evidence status is not a text property but a claim-evidence relation: text-only rankers and defenses cannot separate honest detailed content from fabricated detail, creating an identifiability gap. We audit this gap with an evidence-paired benchmark (50 e-commerce queries, 1,950 cases) and a claim-level reranker, GroundedGEO, that penalizes query-relevant claims lacking support in a supplied packet. Matched rich variants control format and volume; packet twins add attestations at fixed text, while thinned packets withdraw them. On the frozen listwise ranker Qwen2.5-7B, unsupported-rich variants show significant normalized rank gain over clean candidates (+0.065 to +0.092 across claim profiles, Holm-corrected), while supported and neutral controls do not; the effect is model-dependent (marginal on MiMo-v2.5, absent on GLM-5.3-Flash). On a frozen pointwise scorer, oracle evidence labels cut the unsupported-rich top-3 rate from 0.65 to 0.43 (laundering from 0.61 to 0.39) at lambda=40 with zero false suppression; packet twins restore the original rates without changing text. Against a 370-claim human gold, all tested automatic judges fail the preregistered reliability gate, although the best local judge retains 79-100% of oracle suppression with zero measured false suppression on protected arms. Separately, stripping attestation coverage increases false suppression by 0.307. These diagnostic effects identify two limits on the evidence channel: label quality and packet coverage. They do not validate an automatic defense, and interpretation of the adverse human-gold arm remains pending adjudication. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.AI). Categorie arXiv: Information Retrieval (cs.IR) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.25189 — 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 2026-10-08T13:18:38.215Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).