접착제

  • 설명 :

일반 언어 이해 평가 벤치마크( https://gluebenchmark.com/ )인 GLUE는 자연어 이해 시스템을 교육, 평가 및 분석하기 위한 리소스 모음입니다.

접착제/콜라(기본 구성)

  • 구성 설명 : The Corpus of Linguistic Acceptability는 언어 이론에 관한 책과 저널 기사에서 가져온 영어 수용성 판단으로 구성됩니다. 각 예는 문법적인 영어 문장인지 여부가 주석으로 표시된 일련의 단어입니다.

  • 홈페이지 : https://nyu-mll.github.io/CoLA/

  • 다운로드 크기 : 368.14 KiB

  • 데이터 세트 크기 : 965.49 KiB

  • 분할 :

나뉘다
'test' 1,063
'train' 8,551
'validation' 1,043
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'sentence': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 클래스 레이블 int64
문장 텍스트
  • 인용 :
@article{warstadt2018neural,
  title={Neural Network Acceptability Judgments},
  author={Warstadt, Alex and Singh, Amanpreet and Bowman, Samuel R},
  journal={arXiv preprint arXiv:1805.12471},
  year={2018}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/sst2

  • 구성 설명 : Stanford Sentiment Treebank는 영화 리뷰의 문장과 감정에 대한 사람의 주석으로 구성됩니다. 과제는 주어진 문장의 감정을 예측하는 것입니다. 양방향(긍정적/부정적) 클래스 분할을 사용하고 문장 수준 레이블만 사용합니다.

  • 홈페이지 : https://nlp.stanford.edu/sentiment/index.html

  • 다운로드 크기 : 7.09 MiB

  • 데이터 세트 크기 : 7.22 MiB

  • 분할 :

나뉘다
'test' 1,821
'train' 67,349
'validation' 872
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'sentence': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 클래스 레이블 int64
문장 텍스트
  • 인용 :
@inproceedings{socher2013recursive,
  title={Recursive deep models for semantic compositionality over a sentiment treebank},
  author={Socher, Richard and Perelygin, Alex and Wu, Jean and Chuang, Jason and Manning, Christopher D and Ng, Andrew and Potts, Christopher},
  booktitle={Proceedings of the 2013 conference on empirical methods in natural language processing},
  pages={1631--1642},
  year={2013}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/MRPC

  • 구성 설명 : Microsoft Research Paraphrase Corpus(Dolan & Brockett, 2005)는 온라인 뉴스 소스에서 자동으로 추출된 문장 쌍 모음으로, 쌍의 문장이 의미론적으로 동일한지 여부에 대한 사람의 주석이 있습니다.

  • 홈페이지 : https://www.microsoft.com/en-us/download/details.aspx?id=52398

  • 다운로드 크기 : 1.43 MiB

  • 데이터 세트 크기 : 1.74 MiB

  • 분할 :

나뉘다
'test' 1,725
'train' 3,668
'validation' 408
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'sentence1': Text(shape=(), dtype=string),
    'sentence2': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 클래스 레이블 int64
문장1 텍스트
문장2 텍스트
  • 인용 :
@inproceedings{dolan2005automatically,
  title={Automatically constructing a corpus of sentential paraphrases},
  author={Dolan, William B and Brockett, Chris},
  booktitle={Proceedings of the Third International Workshop on Paraphrasing (IWP2005)},
  year={2005}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/qqp

  • 구성 설명 : Quora Question Pairs2 데이터 세트는 커뮤니티 질문 답변 웹사이트 Quora의 질문 쌍 모음입니다. 작업은 한 쌍의 질문이 의미론적으로 동일한지 여부를 결정하는 것입니다.

  • 홈페이지 : https://data.quora.com/First-Quora-Dataset-Release-Question-Pairs

  • 다운로드 크기 : 39.76 MiB

  • 데이터 세트 크기 : 150.37 MiB

  • 분할 :

나뉘다
'test' 390,965
'train' 363,846
'validation' 40,430
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'question1': Text(shape=(), dtype=string),
    'question2': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 클래스 레이블 int64
질문 1 텍스트
질문2 텍스트
  • 인용 :
@online{WinNT,
  author = {Iyer, Shankar and Dandekar, Nikhil and Csernai, Kornel},
  title = {First Quora Dataset Release: Question Pairs},
  year = 2017,
  url = {https://data.quora.com/First-Quora-Dataset-Release-Question-Pairs},
  urldate = {2019-04-03}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/stsb

  • 구성 설명 : Semantic Textual Similarity Benchmark(Cer et al., 2017)는 뉴스 헤드라인, 동영상 및 이미지 캡션, 자연어 추론 데이터에서 가져온 문장 쌍 모음입니다. 각 쌍은 0에서 5까지의 유사성 점수로 사람이 주석을 달았습니다.

  • 홈페이지 : http://ixa2.si.ehu.es/stswiki/index.php/STSbenchmark

  • 다운로드 크기 : 784.05 KiB

  • 데이터 세트 크기 : 1.58 MiB

  • 분할 :

나뉘다
'test' 1,379
'train' 5,749
'validation' 1,500
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': float32,
    'sentence1': Text(shape=(), dtype=string),
    'sentence2': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 텐서 float32
문장1 텍스트
문장2 텍스트
  • 인용 :
@article{cer2017semeval,
  title={Semeval-2017 task 1: Semantic textual similarity-multilingual and cross-lingual focused evaluation},
  author={Cer, Daniel and Diab, Mona and Agirre, Eneko and Lopez-Gazpio, Inigo and Specia, Lucia},
  journal={arXiv preprint arXiv:1708.00055},
  year={2017}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/mnli

  • 구성 설명 : Multi-Genre Natural Language Inference Corpus는 텍스트 수반 주석이 포함된 문장 쌍의 크라우드 소싱 모음입니다. 전제 문장과 가설 문장이 주어졌을 때, 과제는 전제가 가설을 수반하는지(entailment), 가설과 모순되는지(모순), 둘 다 없는지(중립)를 예측하는 것입니다. 전제 문장은 필사된 연설, 허구 및 정부 보고서를 포함하여 10개의 다른 출처에서 수집됩니다. 저자로부터 개인 레이블을 얻은 표준 테스트 세트를 사용하고 일치(도메인 내) 및 불일치(도메인 간) 섹션 모두에서 평가합니다. 또한 SNLI 코퍼스를 보조 훈련 데이터의 550k 예제로 사용하고 권장합니다.

  • 홈페이지 : http://www.nyu.edu/projects/bowman/multinli/

  • 다운로드 크기 : 298.29 MiB

  • 데이터 세트 크기 : 100.56 MiB

  • 분할 :

나뉘다
'test_matched' 9,796
'test_mismatched' 9,847
'train' 392,702
'validation_matched' 9,815
'validation_mismatched' 9,832
  • 기능 구조 :
FeaturesDict({
    'hypothesis': Text(shape=(), dtype=string),
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=3),
    'premise': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
가설 텍스트
아이디엑스 텐서 int32
상표 클래스 레이블 int64
전제 텍스트
  • 인용 :
@InProceedings{N18-1101,
  author = "Williams, Adina
            and Nangia, Nikita
            and Bowman, Samuel",
  title = "A Broad-Coverage Challenge Corpus for
           Sentence Understanding through Inference",
  booktitle = "Proceedings of the 2018 Conference of
               the North American Chapter of the
               Association for Computational Linguistics:
               Human Language Technologies, Volume 1 (Long
               Papers)",
  year = "2018",
  publisher = "Association for Computational Linguistics",
  pages = "1112--1122",
  location = "New Orleans, Louisiana",
  url = "http://aclweb.org/anthology/N18-1101"
}
@article{bowman2015large,
  title={A large annotated corpus for learning natural language inference},
  author={Bowman, Samuel R and Angeli, Gabor and Potts, Christopher and Manning, Christopher D},
  journal={arXiv preprint arXiv:1508.05326},
  year={2015}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/mnli_mismatched

  • 구성 설명 : MNLI에서 일치하지 않는 유효성 검사 및 테스트 분할. 추가 정보는 "mnli" BuilderConfig를 참조하십시오.

  • 홈페이지 : http://www.nyu.edu/projects/bowman/multinli/

  • 다운로드 크기 : 298.29 MiB

  • 데이터 세트 크기 : 4.79 MiB

  • 분할 :

나뉘다
'test' 9,847
'validation' 9,832
  • 기능 구조 :
FeaturesDict({
    'hypothesis': Text(shape=(), dtype=string),
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=3),
    'premise': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
가설 텍스트
아이디엑스 텐서 int32
상표 클래스 레이블 int64
전제 텍스트
  • 인용 :
@InProceedings{N18-1101,
  author = "Williams, Adina
            and Nangia, Nikita
            and Bowman, Samuel",
  title = "A Broad-Coverage Challenge Corpus for
           Sentence Understanding through Inference",
  booktitle = "Proceedings of the 2018 Conference of
               the North American Chapter of the
               Association for Computational Linguistics:
               Human Language Technologies, Volume 1 (Long
               Papers)",
  year = "2018",
  publisher = "Association for Computational Linguistics",
  pages = "1112--1122",
  location = "New Orleans, Louisiana",
  url = "http://aclweb.org/anthology/N18-1101"
}
@article{bowman2015large,
  title={A large annotated corpus for learning natural language inference},
  author={Bowman, Samuel R and Angeli, Gabor and Potts, Christopher and Manning, Christopher D},
  journal={arXiv preprint arXiv:1508.05326},
  year={2015}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/mnli_matched

  • 구성 설명 : MNLI에서 일치하는 유효성 검사 및 테스트 분할입니다. 추가 정보는 "mnli" BuilderConfig를 참조하십시오.

  • 홈페이지 : http://www.nyu.edu/projects/bowman/multinli/

  • 다운로드 크기 : 298.29 MiB

  • 데이터 세트 크기 : 4.58 MiB

  • 분할 :

나뉘다
'test' 9,796
'validation' 9,815
  • 기능 구조 :
FeaturesDict({
    'hypothesis': Text(shape=(), dtype=string),
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=3),
    'premise': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
가설 텍스트
아이디엑스 텐서 int32
상표 클래스 레이블 int64
전제 텍스트
  • 인용 :
@InProceedings{N18-1101,
  author = "Williams, Adina
            and Nangia, Nikita
            and Bowman, Samuel",
  title = "A Broad-Coverage Challenge Corpus for
           Sentence Understanding through Inference",
  booktitle = "Proceedings of the 2018 Conference of
               the North American Chapter of the
               Association for Computational Linguistics:
               Human Language Technologies, Volume 1 (Long
               Papers)",
  year = "2018",
  publisher = "Association for Computational Linguistics",
  pages = "1112--1122",
  location = "New Orleans, Louisiana",
  url = "http://aclweb.org/anthology/N18-1101"
}
@article{bowman2015large,
  title={A large annotated corpus for learning natural language inference},
  author={Bowman, Samuel R and Angeli, Gabor and Potts, Christopher and Manning, Christopher D},
  journal={arXiv preprint arXiv:1508.05326},
  year={2015}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/qnli

  • Config description : Stanford Question Answering Dataset은 문단의 문장 중 하나(Wikipedia에서 가져옴)에 해당 질문에 대한 답변(주석자가 작성)이 포함된 질문-문단 쌍으로 구성된 질문 답변 데이터 세트입니다. 각 질문과 해당 문맥의 각 문장 사이에 쌍을 형성하고 질문과 문맥 문장 사이의 어휘 중복이 적은 쌍을 필터링하여 작업을 문장 쌍 분류로 변환합니다. 작업은 문맥 문장에 질문에 대한 답변이 포함되어 있는지 확인하는 것입니다. 원래 작업의 이 수정된 버전은 모델이 정확한 답을 선택해야 한다는 요구 사항을 제거하지만 답이 항상 입력에 존재하고 어휘 중복이 신뢰할 수 있는 단서라는 단순화된 가정도 제거합니다.

  • 홈페이지 : https://rajpurkar.github.io/SQuAD-explorer/

  • 다운로드 크기 : 10.14 MiB

  • 데이터 세트 크기 : 32.99 MiB

  • 분할 :

나뉘다
'test' 5,463
'train' 104,743
'validation' 5,463
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'question': Text(shape=(), dtype=string),
    'sentence': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 클래스 레이블 int64
의문 텍스트
문장 텍스트
  • 인용 :
@article{rajpurkar2016squad,
  title={Squad: 100,000+ questions for machine comprehension of text},
  author={Rajpurkar, Pranav and Zhang, Jian and Lopyrev, Konstantin and Liang, Percy},
  journal={arXiv preprint arXiv:1606.05250},
  year={2016}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/rte

  • 구성 설명 : RTE(Recognizing Textual Entailment) 데이터 세트는 일련의 연간 텍스트 수반 과제에서 비롯됩니다. RTE1(Dagan et al., 2006), RTE2(Bar Haim et al., 2006), RTE3(Giampiccolo et al., 2007) 및 RTE5(Bentivogli et al., 2009)의 데이터를 결합합니다.4 예는 다음과 같습니다. 뉴스 및 Wikipedia 텍스트를 기반으로 구성되었습니다. 우리는 모든 데이터 세트를 2개 클래스 분할로 변환합니다. 여기서 3개 클래스 데이터 세트의 경우 일관성을 위해 중립 및 모순을 수반하지 않음으로 축소합니다.

  • 홈페이지 : https://aclweb.org/aclwiki/Recognizing_Textual_Entailment

  • 다운로드 크기 : 680.81 KiB

  • 데이터 세트 크기 : 2.15 MiB

  • 분할 :

나뉘다
'test' 3,000
'train' 2,490
'validation' 277
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'sentence1': Text(shape=(), dtype=string),
    'sentence2': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 클래스 레이블 int64
문장1 텍스트
문장2 텍스트
  • 인용 :
@inproceedings{dagan2005pascal,
  title={The PASCAL recognising textual entailment challenge},
  author={Dagan, Ido and Glickman, Oren and Magnini, Bernardo},
  booktitle={Machine Learning Challenges Workshop},
  pages={177--190},
  year={2005},
  organization={Springer}
}
@inproceedings{bar2006second,
  title={The second pascal recognising textual entailment challenge},
  author={Bar-Haim, Roy and Dagan, Ido and Dolan, Bill and Ferro, Lisa and Giampiccolo, Danilo and Magnini, Bernardo and Szpektor, Idan},
  booktitle={Proceedings of the second PASCAL challenges workshop on recognising textual entailment},
  volume={6},
  number={1},
  pages={6--4},
  year={2006},
  organization={Venice}
}
@inproceedings{giampiccolo2007third,
  title={The third pascal recognizing textual entailment challenge},
  author={Giampiccolo, Danilo and Magnini, Bernardo and Dagan, Ido and Dolan, Bill},
  booktitle={Proceedings of the ACL-PASCAL workshop on textual entailment and paraphrasing},
  pages={1--9},
  year={2007},
  organization={Association for Computational Linguistics}
}
@inproceedings{bentivogli2009fifth,
  title={The Fifth PASCAL Recognizing Textual Entailment Challenge.},
  author={Bentivogli, Luisa and Clark, Peter and Dagan, Ido and Giampiccolo, Danilo},
  booktitle={TAC},
  year={2009}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

접착제/wnli

  • 구성 설명 : Winograd 스키마 챌린지(Levesque et al., 2011)는 시스템이 대명사가 포함된 문장을 읽고 선택 목록에서 해당 대명사의 지시어를 선택해야 하는 독해 작업입니다. 예제는 간단한 통계 방법을 저지하기 위해 수동으로 구성됩니다. 각 예제는 문장의 단일 단어 또는 구에서 제공하는 문맥 정보에 따라 달라집니다. 문제를 문장 쌍 분류로 변환하기 위해 모호한 대명사를 가능한 각 지시 대상으로 대체하여 문장 쌍을 구성합니다. 과제는 대명사가 대체된 문장이 원래 문장에 수반되는지 예측하는 것입니다. 우리는 원본 말뭉치의 저자가 비공개로 공유한 소설책에서 파생된 새로운 사례로 구성된 작은 평가 세트를 사용합니다. 포함된 트레이닝 세트는 두 클래스 사이에서 균형을 이루는 반면 테스트 세트는 두 클래스 사이에서 불균형합니다(65%는 수반되지 않음). 또한 데이터 특성으로 인해 개발 세트는 적대적입니다. 때때로 가설은 훈련과 개발 예제 간에 공유되므로 모델이 훈련 예제를 기억하면 해당 개발 세트 예제에서 잘못된 레이블을 예측합니다. QNLI와 마찬가지로 각 예제는 개별적으로 평가되므로 이 작업에 대한 모델의 점수와 변환되지 않은 원래 작업에 대한 점수 사이에 체계적인 대응이 없습니다. 변환된 데이터 세트 WNLI(Winograd NLI)라고 합니다.

  • 홈페이지 : https://cs.nyu.edu/faculty/davise/papers/WinogradSchemas/WS.html

  • 다운로드 크기 : 28.32 KiB

  • 데이터 세트 크기 : 198.88 KiB

  • 분할 :

나뉘다
'test' 146
'train' 635
'validation' 71
  • 기능 구조 :
FeaturesDict({
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'sentence1': Text(shape=(), dtype=string),
    'sentence2': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
아이디엑스 텐서 int32
상표 클래스 레이블 int64
문장1 텍스트
문장2 텍스트
  • 인용 :
@inproceedings{levesque2012winograd,
  title={The winograd schema challenge},
  author={Levesque, Hector and Davis, Ernest and Morgenstern, Leora},
  booktitle={Thirteenth International Conference on the Principles of Knowledge Representation and Reasoning},
  year={2012}
}
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.

풀/도끼

  • 구성 설명 : 광범위한 언어 현상에 대한 시스템 성능의 세분화된 분석을 위해 수동으로 선별된 평가 데이터 세트입니다. 이 데이터 세트는 자연어 추론(NLI) 문제를 통해 문장 이해도를 평가합니다. MulitNLI에서 훈련된 모델을 사용하여 이 데이터 세트에 대한 예측을 생성합니다.

  • 홈페이지 : https://gluebenchmark.com/diagnostics

  • 다운로드 크기 : 217.05 KiB

  • 데이터 세트 크기 : 299.16 KiB

  • 분할 :

나뉘다
'test' 1,104
  • 기능 구조 :
FeaturesDict({
    'hypothesis': Text(shape=(), dtype=string),
    'idx': int32,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=3),
    'premise': Text(shape=(), dtype=string),
})
  • 기능 문서 :
특징 수업 모양 D타입 설명
풍모Dict
가설 텍스트
아이디엑스 텐서 int32
상표 클래스 레이블 int64
전제 텍스트
  • 인용 :
@inproceedings{wang2019glue,
  title={ {GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.},
  note={In the Proceedings of ICLR.},
  year={2019}
}

Note that each GLUE dataset has its own citation. Please see the source to see
the correct citation for each contained dataset.