cos_e

  • বর্ণনা :

সাধারণ জ্ঞানের ব্যাখ্যা (CoS-E) ভাষা মডেলকে স্বয়ংক্রিয়ভাবে ব্যাখ্যা তৈরি করার জন্য প্রশিক্ষণের অনুমতি দেয় যা প্রশিক্ষণের সময় ব্যবহার করা যেতে পারে এবং একটি উপন্যাস কমনসেন্স অটো-জেনারেটেড ব্যাখ্যা (CAGE) ফ্রেমওয়ার্কের মধ্যে অনুমান করা যেতে পারে।

বিভক্ত উদাহরণ
'train' 9,741
'validation' 1,221
  • বৈশিষ্ট্য গঠন :
FeaturesDict({
    'abstractive_explanation': Text(shape=(), dtype=string),
    'answer': Text(shape=(), dtype=string),
    'choices': Sequence(Text(shape=(), dtype=string)),
    'extractive_explanation': Text(shape=(), dtype=string),
    'id': Text(shape=(), dtype=string),
    'question': Text(shape=(), dtype=string),
})
  • বৈশিষ্ট্য ডকুমেন্টেশন :
বৈশিষ্ট্য ক্লাস আকৃতি ডিটাইপ বর্ণনা
ফিচারসডিক্ট
বিমূর্ত_ব্যাখ্যা পাঠ্য স্ট্রিং
উত্তর পাঠ্য স্ট্রিং
পছন্দ ক্রম (পাঠ্য) (কোনটিই নয়,) স্ট্রিং
extractive_explanation পাঠ্য স্ট্রিং
আইডি পাঠ্য স্ট্রিং
প্রশ্ন পাঠ্য স্ট্রিং
  • উদ্ধৃতি :
@inproceedings{rajani2019explain,
     title = "Explain Yourself! Leveraging Language models for Commonsense Reasoning",
    author = "Rajani, Nazneen Fatema  and
      McCann, Bryan  and
      Xiong, Caiming  and
      Socher, Richard",
      year="2019",
    booktitle = "Proceedings of the 2019 Conference of the Association for Computational Linguistics (ACL2019)",
    url ="https://arxiv.org/abs/1906.02361"
}