{
  "abstract": "Assessing compound weather risks requires forecasts representing dependence between variables. CLARA (Calibrated Advection-Routing Attention) learns joint Gaussian predictive distributions of five surface variables from station observations alone, without numerical weather prediction or reanalysis; the approximately 28,000-parameter model supports CPU training and prediction. Across six multi-year folds on 96 stations, its lead-mean energy score is 4.9% lower than that of a learned comparator with matched temporal inputs (4.7% with a similar parameter count) and 11-65% lower than those of statistical baselines. Holding marginal variances fixed, removing learned correlations worsens joint negative log-likelihood by 1.0-2.8 nats per station. A covariance-scale estimator, proved consistent under stated assumptions, improves short-lead calibration but over-corrects at long leads. Synthetic interventions show an attention-bias coefficient alone does not measure forecast influence. Retrained in ten regions on six continents, CLARA outperforms persistence in all 60 multi-year region-lead comparisons and a similarly sized learned model in 57 of 60.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.09898",
  "author_total": 2,
  "authors": [
    {
      "name": "Chaeyeon Yi"
    },
    {
      "name": "Yun Am Seo"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/5bqknbbcc6dtp.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.09898",
  "fulltext_url": null,
  "id": "5bqknbbcc6dtp",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.09898",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "physics.ao-ph"
    ],
    [
      "arxiv",
      "stat.AP"
    ],
    [
      "openalex",
      "1902|Meteorological Phenomena and Simulations"
    ]
  ],
  "openalex": "W7220923596",
  "pdf_url": "https://arxiv.org/pdf/2610.09898",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-07",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/5bqknbbcc6dtp.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:37Z",
      "native_id": "2610.09898",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.09898"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "arxiv:physics.ao-ph",
      "subfield": "atmospheric"
    },
    {
      "field": "mathematics",
      "origin": "arxiv:stat.AP",
      "subfield": "statistics"
    }
  ],
  "title": "Learning joint probabilistic weather forecasts from station observations alone",
  "title_text": "Learning joint probabilistic weather forecasts from station observations alone",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:48Z",
  "url": "https://research.draconitic.com/paper/5bqknbbcc6dtp",
  "venue": "arXiv",
  "versions": []
}