{
  "abstract": "Steel property models can perform well on random test splits even when they extrapolate poorly to unfamiliar alloys. SteelAgent couples metallurgical descriptors and machine-learning models with a language-model interface for property prediction, uncertainty reporting and inverse design. We examine how the choice of held-out unit changes the conclusions drawn from SteelBench tensile-strength evaluations. For gradient boosting, pooled out-of-fold $R^2$ fell from 0.859 on a random split to 0.113 when steel families were held out. Empirical coverage of nominal 90% prediction intervals ranged from 91.0% under random splitting to 37.6% on held-out sources. The effect of physics-derived features depended on the evaluation. They increased gradient-boosting $R^2$ by 0.2534 in a 17-family holdout analysis and decreased it by 0.358 in a separate four-grade laboratory analysis. We report group-level errors and a synthetic regression experiment illustrating how calibration can deteriorate on held-out groups. These results motivate reporting the evaluation population alongside a property estimate or compliance score.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2611.06953",
  "author_total": 4,
  "authors": [
    {
      "name": "Aleksandr Volkov"
    },
    {
      "name": "Roman Sultimov"
    },
    {
      "name": "Mikhail Kuzin"
    },
    {
      "name": "Yury Maximov"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/3h8zpz0fyrg2g.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2611.06953",
  "fulltext_url": null,
  "id": "3h8zpz0fyrg2g",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2611.06953",
  "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",
      "cond-mat.mtrl-sci"
    ],
    [
      "arxiv",
      "cs.LG"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://arxiv.org/pdf/2611.06953",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-03",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/3h8zpz0fyrg2g.ris",
  "sources": [
    {
      "first_seen": "2026-11-06T18:21:30Z",
      "native_id": "2611.06953",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2611.06953"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "arxiv:cond-mat.mtrl-sci",
      "subfield": "condensed-matter"
    }
  ],
  "title": "SteelAgent: LLM-Orchestrated, Physics-Informed Steel Property Prediction with Audited Generalization and Calibrated Uncertainty",
  "title_text": "SteelAgent: LLM-Orchestrated, Physics-Informed Steel Property Prediction with Audited Generalization and Calibrated Uncertainty",
  "type": "preprint",
  "updated_at": "2026-11-09T00:20:39Z",
  "url": "https://research.draconitic.com/paper/3h8zpz0fyrg2g",
  "venue": "arXiv",
  "versions": []
}