{
  "abstract": "We present a new strategy for unsupervised anomaly detection at the hardware-based first stage of event processing (Level-1 trigger) of the Large Hadron Collider (LHC). A normalising flow trained exclusively on Standard Model (SM) events provides a teacher anomaly score based on its exact negative log-likelihood, which is distilled into a compact neural network suitable for field-programmable gate array (FPGA) deployment. The teacher reaches state-of-the-art performance on a benchmark dataset, using a rigorous p-value-based definition of the anomaly score. A simple two-hidden-layer student reproduces the teacher performance across four beyond-the-SM benchmarks to within 0.1 percentage points in area under the receiver operating characteristic curve (AUC), while achieving a compression factor of approximately 325 times. Quantisation-aware training with PQuantML reduces the model to 8-bit precision with AUC changes below 0.06 percentage points. Compilation to Verilog firmware with Alkaid yields FPGA designs that require neither digital signal processors nor block random-access memory and achieve latencies of 27-52 ns. This modular pipeline decouples the expressiveness of the anomaly-detection model from real-time hardware constraints, enabling complex architectures to be used for model-agnostic searches for new physics at the full LHC collision rate of 40 MHz.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.15295",
  "author_total": 7,
  "authors": [
    {
      "name": "Jaiman Abson"
    },
    {
      "name": "Florencia Canelli"
    },
    {
      "name": "Valentina Guglielmi"
    },
    {
      "name": "Roope Oskari Niemi"
    },
    {
      "name": "Maurizio Pierini"
    },
    {
      "name": "Chang Sun"
    },
    {
      "name": "Francesco Vaselli"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/eszvhp8zzy835.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.15295",
  "fulltext_url": null,
  "id": "eszvhp8zzy835",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.15295",
  "language": null,
  "licence": "CC-BY-SA-4.0",
  "licence_label": "CC BY-SA 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by-sa/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "hep-ex"
    ],
    [
      "openalex",
      "1702|Anomaly Detection Techniques and Applications"
    ]
  ],
  "openalex": "W7213369965",
  "pdf_url": "https://arxiv.org/pdf/2609.15295",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-17",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/eszvhp8zzy835.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:22:19Z",
      "native_id": "2609.15295",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.15295"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "arxiv:hep-ex",
      "subfield": "high-energy"
    }
  ],
  "title": "Knowledge Distillation of a Normalising Flow for Real-Time Anomaly Detection at LHC Level-1 Trigger",
  "title_text": "Knowledge Distillation of a Normalising Flow for Real-Time Anomaly Detection at LHC Level-1 Trigger",
  "type": "preprint",
  "updated_at": "2026-10-30T12:22:19Z",
  "url": "https://research.draconitic.com/paper/eszvhp8zzy835",
  "venue": "arXiv",
  "versions": []
}