{
  "abstract": "PaECTER is an open-source document-level encoder specific to patents. We fine-tune BERT for Patents with examiner-added citation information to generate numerical representations for patent documents. PaECTER performs better in similarity tasks than current state-of-the-art models used in the patent domain. More specifically, our model outperforms the patent specific pre-trained language model (BERT for Patents) and general-purpose text embedding models (e.g., E5, GTE, and BGE) on our patent citation prediction test dataset on different rank evaluation metrics. PaECTER predicts at least one most similar patent at a rank of 1.32 on average when compared against 25 irrelevant patents. Numerical representations generated by PaECTER from patent text can be used for downstream tasks such as classification, tracing knowledge flows, or semantic similarity search. Semantic similarity search is especially relevant in the context of prior art search for both inventors and patent examiners.",
  "abstract_shown": true,
  "affiliations": [
    "Innovation & Entrepreneurship Research, Max Planck Institute for Innovation and Competition, Munich, Germany",
    "Ludwig Maximilian University of Munich, Munich, Germany",
    "Centre for Economic Policy Research (CEPR), London, United Kingdom"
  ],
  "article_type": "Research Article",
  "arxiv": null,
  "author_total": 5,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Mainak Ghosh"
    },
    {
      "affs": [
        0
      ],
      "name": "Michael E. Rose",
      "orcid": "0000-0002-4128-4236"
    },
    {
      "affs": [
        0
      ],
      "name": "Sebastian Erhardt",
      "orcid": "0000-0002-2933-6451"
    },
    {
      "affs": [
        0
      ],
      "name": "Erik Buunk"
    },
    {
      "affs": [
        0,
        1,
        2
      ],
      "name": "Dietmar Harhoff"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/1csawmpmg622n.bib",
  "cited_by": 0,
  "doi": "10.1371/journal.pone.0358188",
  "fulltext_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC13712717/",
  "id": "1csawmpmg622n",
  "issns": [
    "1932-6203"
  ],
  "landing_url": "https://doi.org/10.1371/journal.pone.0358188",
  "language": "en",
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "heading",
      "Algorithms"
    ],
    [
      "heading",
      "Applied Mathematics"
    ],
    [
      "heading",
      "Artificial Intelligence"
    ],
    [
      "heading",
      "Biology and Life Sciences"
    ],
    [
      "heading",
      "Citation Analysis"
    ],
    [
      "heading",
      "Cognitive Psychology"
    ],
    [
      "heading",
      "Cognitive Science"
    ],
    [
      "heading",
      "Commercial Law"
    ],
    [
      "heading",
      "Computer and Information Sciences"
    ],
    [
      "heading",
      "Engineering and Technology"
    ],
    [
      "heading",
      "Grammar"
    ],
    [
      "heading",
      "Language"
    ],
    [
      "heading",
      "Language Acquisition"
    ],
    [
      "heading",
      "Law and Legal Sciences"
    ],
    [
      "heading",
      "Linguistics"
    ],
    [
      "heading",
      "Machine Learning"
    ],
    [
      "heading",
      "Mathematics"
    ],
    [
      "heading",
      "Neuroscience"
    ],
    [
      "heading",
      "Patents"
    ],
    [
      "heading",
      "Phonology"
    ],
    [
      "heading",
      "Physical Sciences"
    ],
    [
      "heading",
      "Programming Languages"
    ],
    [
      "heading",
      "Psychology"
    ],
    [
      "heading",
      "Research Article"
    ],
    [
      "heading",
      "Research Assessment"
    ],
    [
      "heading",
      "Research and Analysis Methods"
    ],
    [
      "heading",
      "Simulation and Modeling"
    ],
    [
      "heading",
      "Social Sciences"
    ],
    [
      "heading",
      "Software Engineering"
    ],
    [
      "heading",
      "Vocabulary"
    ],
    [
      "openalex",
      "1702|Topic Modeling"
    ],
    [
      "plos",
      "/Biology and life sciences/Neuroscience/Cognitive science/Cognitive psychology/Language"
    ],
    [
      "plos",
      "/Biology and life sciences/Psychology/Cognitive psychology/Language"
    ],
    [
      "plos",
      "/Computer and information sciences/Artificial intelligence/Machine learning"
    ],
    [
      "plos",
      "/Computer and information sciences/Software engineering/Programming languages"
    ],
    [
      "plos",
      "/Engineering and technology/Software engineering/Programming languages"
    ],
    [
      "plos",
      "/Physical sciences/Mathematics/Applied mathematics/Algorithms"
    ],
    [
      "plos",
      "/Research and analysis methods/Research assessment/Citation analysis"
    ],
    [
      "plos",
      "/Research and analysis methods/Simulation and modeling/Algorithms"
    ],
    [
      "plos",
      "/Social sciences/Law and legal sciences/Commercial law/Patents"
    ],
    [
      "plos",
      "/Social sciences/Linguistics/Grammar/Phonology/Vocabulary"
    ],
    [
      "plos",
      "/Social sciences/Linguistics/Language acquisition"
    ],
    [
      "plos",
      "/Social sciences/Psychology/Cognitive psychology/Language"
    ]
  ],
  "openalex": "W4415308427",
  "pdf_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0358188&type=printable",
  "pmcid": "PMC13712717",
  "pmid": "42945360",
  "published": "2026-10-27",
  "publisher": "Public Library of Science",
  "ris": "https://research.draconitic.com/paper/1csawmpmg622n.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:17Z",
      "native_id": "10.1371/journal.pone.0358188",
      "source": "plos",
      "url": "https://doi.org/10.1371/journal.pone.0358188"
    },
    {
      "first_seen": "2026-10-31T12:23:14Z",
      "native_id": "PMC13712717",
      "source": "pmc",
      "url": "https://doi.org/10.1371/journal.pone.0358188"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "plos:neuroscience",
      "subfield": "neuroscience"
    },
    {
      "field": "mathematics",
      "origin": "heading:Mathematics",
      "subfield": "general"
    }
  ],
  "title": "PaECTER: Patent-level representation learning using citation-informed transformers",
  "title_text": "PaECTER: Patent-level representation learning using citation-informed transformers",
  "type": "peer-reviewed",
  "updated_at": "2026-10-31T12:23:14Z",
  "url": "https://research.draconitic.com/paper/1csawmpmg622n",
  "venue": "PLOS One",
  "versions": []
}