Metraipolis › archive › dc_d9eb511669d6
EviCast: Forecasting Retrieval Gains Through Shared Response Calibration — arXiv:2611.02387
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02387
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:23:39.071Z
- SHA-256 of the archived text
d639b415d29303a11021322a2e05e9f62d1f5db4ffde93c11f4a54cc4f658c19- Stable URL of this page
https://metraipolis.com/documento/dc_d9eb511669d6
How to cite
EviCast: Forecasting Retrieval Gains Through Shared Response Calibration — arXiv:2611.02387 — arXiv (en), https://arxiv.org/abs/2611.02387 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d9eb511669d6 (Metraipolis archive dc_d9eb511669d6, sha256 d639b415d29303a11021322a2e05e9f62d1f5db4ffde93c11f4a54cc4f658c19)
BibTeX
@misc{dcd9eb511669d6,
title = {EviCast: Forecasting Retrieval Gains Through Shared Response Calibration — arXiv:2611.02387},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02387},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d9eb511669d6; sha256 d639b415d29303a11021322a2e05e9f62d1f5db4ffde93c11f4a54cc4f658c19},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02387. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Retrieval-augmented generation (RAG) extends large language models with external evidence, yet relevant evidence does not ensure that a generator can use it. Indiscriminate retrieval can degrade task performance while adding retrieval and long-context inference costs. Existing methods neither determine before large-scale execution whether a generator, task distribution, and retriever configuration is worth deploying nor diagnose multiple candidate retrievers under a shared budget. We introduce EviCast, a framework that forecasts retrieval value through interventional evidence-state decomposition. EviCast measures a generator's response to controlled evidence states once and combines this shared profile with the offline evidence-state distribution of each candidate retriever, yielding the end-to-end net utility and its 95% lower confidence bound without executing every RAG configuration. A state-refinement analysis characterizes when its binary and topology-aware instantiations are exact and when shared calibration is statistically more efficient than per-retriever calibration. In frozen evaluations, EviCast correctly decides all 14 HotpotQA retrieval configurations with a forecast error of 2.75 F1 points while reducing diagnostic generation calls by 80%, and it attains the lowest forecast error among seven predictors in the low-budget regime. Its topology-aware instantiation, EviCast-T, captures the complementarity between terminal and bridge evidence, correctly deciding all 16 configurations on MuSiQue and HotpotQA and 23 of 24 after prospective transfer to a third dataset. These results show that retrieval deployment depends on both evidence availability and the generator's ability to convert evidence structure into stable net utility. Code is available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.02387 — 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 2026-11-03T18:10:07.964Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).