November 2, 2019 Conference paper Open Access

Macé, Valentin; Servan, Christophe

JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/7c30fb05-d597-49ac-86e0-ca27f7dd1351/IWSLT2019_paper_20.pdf"
      }, 
      "checksum": "md5:3a4a4614e5f77e181163dafd99370339", 
      "bucket": "7c30fb05-d597-49ac-86e0-ca27f7dd1351", 
      "key": "IWSLT2019_paper_20.pdf", 
      "type": "pdf", 
      "size": 168252
    }
  ], 
  "owners": [
    50447
  ], 
  "doi": "10.5281/zenodo.3525020", 
  "stats": {
    "version_unique_downloads": 79.0, 
    "unique_views": 111.0, 
    "views": 119.0, 
    "version_views": 119.0, 
    "unique_downloads": 79.0, 
    "version_unique_views": 111.0, 
    "volume": 14469672.0, 
    "version_downloads": 86.0, 
    "downloads": 86.0, 
    "version_volume": 14469672.0
  }, 
  "links": {
    "doi": "https://doi.org/10.5281/zenodo.3525020", 
    "conceptdoi": "https://doi.org/10.5281/zenodo.3525019", 
    "bucket": "https://zenodo.org/api/files/7c30fb05-d597-49ac-86e0-ca27f7dd1351", 
    "conceptbadge": "https://zenodo.org/badge/doi/10.5281/zenodo.3525019.svg", 
    "html": "https://zenodo.org/record/3525020", 
    "latest_html": "https://zenodo.org/record/3525020", 
    "badge": "https://zenodo.org/badge/doi/10.5281/zenodo.3525020.svg", 
    "latest": "https://zenodo.org/api/records/3525020"
  }, 
  "conceptdoi": "10.5281/zenodo.3525019", 
  "created": "2019-11-01T13:14:34.920169+00:00", 
  "updated": "2020-01-20T16:11:45.016805+00:00", 
  "conceptrecid": "3525019", 
  "revision": 2, 
  "id": 3525020, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.5281/zenodo.3525020", 
    "description": "<p>In Machine Translation, considering the document as a whole can help to resolve ambiguities and inconsistencies. In this paper, we propose a simple yet promising approach to add contextual information in Neural Machine Translation. We present a method to add source context that capture the whole document with accurate boundaries, taking every word into account. We provide this additional information to a Transformer model and study the impact of our method on three language pairs. The proposed approach obtains promising results in the English-German, English-French and French-English document-level translation tasks. We observe interesting cross-sentential behaviors where the model learns to use document-level information to improve translation coherence.</p>", 
    "language": "eng", 
    "title": "Using Whole Document Context in Neural Machine Translation", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "3525019"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "3525020"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "iwslt2019"
      }
    ], 
    "publication_date": "2019-11-02", 
    "creators": [
      {
        "affiliation": "QWANT RESEARCH - 7 Rue Spontini, 75116 Paris, France", 
        "name": "Mac\u00e9, Valentin"
      }, 
      {
        "affiliation": "QWANT RESEARCH - 7 Rue Spontini, 75116 Paris, France", 
        "name": "Servan, Christophe"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "subtype": "conferencepaper", 
      "type": "publication", 
      "title": "Conference paper"
    }, 
    "related_identifiers": [
      {
        "scheme": "doi", 
        "identifier": "10.5281/zenodo.3525019", 
        "relation": "isVersionOf"
      }
    ]
  }
}