{
  "abstract": "Knowledge distillation for jet taggers is usually cast as compression toward a near-oracle teacher. We instead train a small reliability-gated correction layer jointly over two frozen teachers --- a GN2-type ATLAS flavour tagger (3 classes) and CMS ParT (10 classes) --- reading only their frozen penultimate embeddings and logits, so it adds negligible marginal cost. A single per-sample reliability signal reweights the distillation loss, drives a prototype-based repair of the teachers' errors, and anchors the inference-time gate that mixes teacher and student logits. Zero-initialised, the 82\\,k-parameter student starts as a bit-exact copy of the teacher --- a provable invariant on the training distribution --- and departs only where the data support it. Under a statistically-grounded evaluation, with binomial, Poisson and paired uncertainties on every table, it reproduces both teachers at $\\geq\\!99\\%$ agreement and matches their overall accuracy while redistributing it toward their weakest classes: the JetClass per-class accuracy of $H\\!\\to\\!4q$, QCD and $H\\!\\to\\!bb$ rises by $+0.8$, $+0.4$ and $+0.2$ percentage points ($|z|>7$), offset by comparable drops elsewhere. Oracle heads on the frozen embedding recover a budget consistent with zero in both domains, and a conformal-risk-control analysis certifies the false-accept rate. This full re-evaluation supersedes an earlier, incorrectly-evaluated set of numbers; code and models are available at https://github.com/sysu17363020/JetCoRD.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2607.15531",
  "author_total": 5,
  "authors": [
    {
      "name": "Liu-Long Gao"
    },
    {
      "name": "Zheng-Kun Huang"
    },
    {
      "name": "Xiao-Wei Jiang"
    },
    {
      "name": "Jia-Feng Li"
    },
    {
      "name": "Gong-Xing Sun"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/cs1mmnefh5bwb.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2607.15531",
  "fulltext_url": null,
  "id": "cs1mmnefh5bwb",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2607.15531",
  "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",
      "hep-ph"
    ],
    [
      "openalex",
      "3106|Particle physics theoretical and experimental studies"
    ]
  ],
  "openalex": "W7169756643",
  "pdf_url": "https://arxiv.org/pdf/2607.15531",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-03",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/cs1mmnefh5bwb.ris",
  "sources": [
    {
      "first_seen": "2026-11-04T15:21:24Z",
      "native_id": "2607.15531",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2607.15531"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "arxiv:hep-ph",
      "subfield": "high-energy"
    }
  ],
  "title": "JetCoRD: Reliability-Gated Correction of Jet Taggers over Two Frozen Teachers",
  "title_text": "JetCoRD: Reliability-Gated Correction of Jet Taggers over Two Frozen Teachers",
  "type": "preprint",
  "updated_at": "2026-11-06T18:21:01Z",
  "url": "https://research.draconitic.com/paper/cs1mmnefh5bwb",
  "venue": "arXiv",
  "versions": []
}