{
  "abstract": "Evaluating a new model on an existing benchmark is often necessary to understand its behavior before deployment. For modern evaluation frameworks, generating and evaluating a response for all queries can be prohibitively expensive. In practice, responses from previously-evaluated models are often cached -- creating a potential opportunity to use this additional information to decrease the number of queries required to accurately evaluate a new model. In this paper, we introduce an approach for predicting benchmark performance that leverages cached model responses based on the Data Kernel Perspective Space (DKPS), a method for quantifying the relationship between models in the black-box setting. Theoretically, we show that DKPS-based methods are query-efficient under certain conditions. Empirically, we demonstrate that DKPS-based methods achieve the same mean absolute error as baselines with a substantially decreased query budget. We conclude by proposing an offline method for selecting a set of queries that maximizes the goodness-of-fit on reference models, improving prediction accuracy over random query selection.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": "Journal reference: Proceedings of the 43rd International Conference on Machine Learning (ICML 2026), PMLR 306:42691-42706, 2026",
  "arxiv": "2605.07096",
  "author_total": 3,
  "authors": [
    {
      "name": "Hayden Helm"
    },
    {
      "name": "Ben Johnson"
    },
    {
      "name": "Carey Priebe"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/cvf3ehmcyt2c7.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2605.07096",
  "fulltext_url": null,
  "id": "cvf3ehmcyt2c7",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2605.07096",
  "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.AI"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ME"
    ],
    [
      "openalex",
      "1702|Machine Learning and Data Classification"
    ]
  ],
  "openalex": "W7160836797",
  "pdf_url": "https://arxiv.org/pdf/2605.07096",
  "pmcid": null,
  "pmid": null,
  "published": "2026-06-04",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/cvf3ehmcyt2c7.ris",
  "sources": [
    {
      "first_seen": "2026-11-04T15:21:34Z",
      "native_id": "2605.07096",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2605.07096"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ME",
      "subfield": "statistics"
    }
  ],
  "title": "Query-efficient model evaluation using cached responses",
  "title_text": "Query-efficient model evaluation using cached responses",
  "type": "preprint",
  "updated_at": "2026-11-06T18:21:10Z",
  "url": "https://research.draconitic.com/paper/cvf3ehmcyt2c7",
  "venue": "arXiv",
  "versions": []
}