codemeta.json

{
  "@context": "https://doi.org/10.5063/schema/codemeta-2.0",
  "@type": "SoftwareSourceCode",
  "identifier": "Paraphrasis",
  "name": "Paraphrasis",
  "description": "Paraphrasis is a research software tool for the detection and human review of paraphrases in text. It combines sentence and clause segmentation with sentence-embedding models to propose candidate paraphrase pairs, provides a web interface for inspecting, annotating and reviewing those candidates, and supports fine-tuning a specialised sentence-embedding model on domain data.",
  "codeRepository": "https://huggingface.co/spaces/TdelaSelle/Paraphrasis",
  "url": "https://tdelaselle-paraphrasis.hf.space/",
  "issueTracker": "https://huggingface.co/spaces/TdelaSelle/Paraphrasis/discussions",
  "readme": "https://huggingface.co/spaces/TdelaSelle/Paraphrasis/blob/main/README.md",
  "license": "https://spdx.org/licenses/MIT",
  "version": "0.1.0",
  "softwareVersion": "0.1.0",
  "dateCreated": "2026-04-23",
  "dateModified": "2026-06-19",
  "datePublished": "2026-06-19",
  "developmentStatus": "active",
  "applicationCategory": "Natural Language Processing",
  "programmingLanguage": [
    {
      "@type": "ComputerLanguage",
      "name": "Python",
      "version": "3"
    },
    {
      "@type": "ComputerLanguage",
      "name": "HTML"
    }
  ],
  "runtimePlatform": "Docker",
  "operatingSystem": "Linux",
  "softwareRequirements": [
    "Python 3",
    "Streamlit",
    "TODO: complete this list from requirements.txt (pinned versions)"
  ],
  "keywords": [
    "paraphrase detection",
    "paraphrase annotation",
    "semantic textual similarity",
    "sentence embeddings",
    "natural language processing",
    "research software"
  ],
  "author": [
    {
      "@type": "Person",
      "givenName": "Théotime",
      "familyName": "de la Selle",
      "email": "TODO: institutional e-mail",
      "identifier": "TODO: https://orcid.org/0000-0000-0000-0000",
      "affiliation": {
        "@type": "Organization",
        "name": "Institut des Sources Chrétiennes, HiSoMA, CNRS, Lyon, France"
      }
    }
  ],
  "maintainer": {
    "@type": "Person",
    "givenName": "Théotime",
    "familyName": "de la Selle"
  },
  "copyrightHolder": {
    "@type": "Person",
    "givenName": "Théotime",
    "familyName": "de la Selle"
  },
  "copyrightYear": 2026,
  "funder": {
    "@type": "Organization",
    "name": "Agence Nationale de la Recherche (Biblissima+)"
  },
  "funding": "ANR-21-ESRE-0005 (Biblissima+, Observatoire des cultures écrites anciennes, de l'argile à l'imprimé)",
  "referencePublication": {
    "@type": "ScholarlyArticle",
    "name": "From Transcription to Semantic Corpus Analysis: Unsupervised Learning of Sentence Representations for Ancient Languages",
    "author": [
      {
        "@type": "Person",
        "givenName": "Théotime",
        "familyName": "de la Selle"
      }
    ],
    "datePublished": "2026-07-27",
    "url": "https://arxiv.org/abs/2607.24542",
    "identifier": "arXiv:2607.24542"
  },
  "relatedLink": [
    "https://tdelaselle.gitpages.huma-num.fr/paraphrasis/",
    "https://github.com/Tdelaselle/PatriSE",
    "https://huggingface.co/TdelaSelle/PatriLaSE"
  ]
}