ब्रह्मांड_क्यूए

  • विवरण :

Cosmos QA 35.6K समस्याओं का एक बड़े पैमाने का डेटासेट है, जिसके लिए सामान्य ज्ञान-आधारित पठन समझ की आवश्यकता होती है, जिसे बहुविकल्पीय प्रश्नों के रूप में तैयार किया जाता है। यह लोगों के रोजमर्रा के आख्यानों के विविध संग्रह पर पंक्तियों के बीच पढ़ने पर ध्यान केंद्रित करता है, घटनाओं के संभावित कारणों या प्रभावों से संबंधित प्रश्न पूछता है जिसके संदर्भ में सटीक पाठ विस्तार से परे तर्क की आवश्यकता होती है।

विभाजित करना उदाहरण
'test' 6,963
'train' 25,262
'validation' 2,985
  • फ़ीचर संरचना :
FeaturesDict({
    'answer0': Text(shape=(), dtype=string),
    'answer1': Text(shape=(), dtype=string),
    'answer2': Text(shape=(), dtype=string),
    'answer3': Text(shape=(), dtype=string),
    'context': Text(shape=(), dtype=string),
    'id': Text(shape=(), dtype=string),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=4),
    'question': Text(shape=(), dtype=string),
})
  • फ़ीचर दस्तावेज़ीकरण :
विशेषता कक्षा आकार डीटाइप विवरण
विशेषताएं डिक्ट
उत्तर0 मूलपाठ डोरी
उत्तर 1 मूलपाठ डोरी
उत्तर2 मूलपाठ डोरी
उत्तर3 मूलपाठ डोरी
संदर्भ मूलपाठ डोरी
पहचान मूलपाठ डोरी
लेबल क्लासलेबल int64
प्रश्न मूलपाठ डोरी
  • उद्धरण :
@inproceedings{huang-etal-2019-cosmos,
    title = "Cosmos {QA}: Machine Reading Comprehension with Contextual Commonsense Reasoning",
    author = "Huang, Lifu  and
      Le Bras, Ronan  and
      Bhagavatula, Chandra  and
      Choi, Yejin",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)",
    year = "2019",
    url = "https://www.aclweb.org/anthology/D19-1243"
}