{
  "abstract": "Question-answering services built on retrieval-augmented generation (RAG), in which a language model answers from retrieved documents, are inspected continuously and upgraded repeatedly, so their reliability guarantee must survive both. We study federated conformal RAG: nodes holding private corpora score candidate answers with a shared language model and send compressed scores to a hub that returns an answer set; a miss omits the true answer. We formulate monitoring as a sequential test: alarm when misses exceed a certified bound on the miss rate (the allowance). But a conformal certificate covers one frozen configuration at one look fixed in advance, so it gives no anytime-valid alarm; a threshold calibrated for one model certifies nothing about its replacement; and re-certifying each upgrade at full error level compounds failures. We propose Anytime-FC-RAG with three auditable rules: pick each configuration before its fresh calibration, charge every certificate attempt to one trajectory-wide budget $δ_{\\rm cal}$, and fix threshold, allowance and bet before each query. One betting wealth $E_t$, growing in expectation only when misses exceed the allowance, never resets across deployments. Our certified adaptive-refresh theorem shows that, with evidence budget $δ_e$, alarming when $E_t$ reaches $1/δ_e$ has false-alarm probability at most $δ_e + δ_{\\rm cal}$ under a random number of history-selected, evidence-driven refreshes of model, retriever, corpus, score or threshold; $δ_{\\rm cal}$ pays for wrong certificates. With an exact order-statistic certificate the bound is distribution-free, architecture-agnostic and about misses, not per-input coverage or distribution shift as such. On Qwen2.5/MMLU-Pro the alarm separated audited-harmful from benign shifts completely at a 5% budget; in a model swap, only fresh calibration kept a downgrade's miss rate in check.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2605.29139",
  "author_total": 2,
  "authors": [
    {
      "name": "Prasanjit Dubey"
    },
    {
      "name": "Xiaoming Huo"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/6sz0ycwt8m59g.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2605.29139",
  "fulltext_url": null,
  "id": "6sz0ycwt8m59g",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2605.29139",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Adversarial Robustness in Machine Learning"
    ]
  ],
  "openalex": "W7162760278",
  "pdf_url": "https://arxiv.org/pdf/2605.29139",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-29",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/6sz0ycwt8m59g.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:25:41Z",
      "native_id": "2605.29139",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2605.29139"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Certified Adaptive Refresh: Anytime-Valid Monitoring for Federated Conformal RAG",
  "title_text": "Certified Adaptive Refresh: Anytime-Valid Monitoring for Federated Conformal RAG",
  "type": "preprint",
  "updated_at": "2026-10-30T12:25:41Z",
  "url": "https://research.draconitic.com/paper/6sz0ycwt8m59g",
  "venue": "arXiv",
  "versions": []
}