Papers with Code · CC BY-SA 4.0

Sentence Pair Modeling

Language Modelling › Sentence Pair Modeling 21 평가 기록 28 datasets

Comparing two sentences and their relationship based on their internal representation.

개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.

벤치마크

Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.

Dataset기록된 모델 예시MetricPaperCode

Annotated corpus for semantic similarity of clinical trial outcomes (expanded corpus)

BioBERT (pre-trained on PubMed abstracts + PMC, fine-tuned on "Annotated corpus for semantic similarity of clinical trial outcomes, expanded corpus")

F1: 93.38Precision: 92.98Recall: 93.85

논문 -

Annotated corpus for semantic similarity of clinical trial outcomes (original corpus)

BioBERT (pre-trained on PubMed abstracts + PMC, fine-tuned on "Annotated corpus for semantic similarity of clinical trial outcomes, original corpus")

F1: 89.75Precision: 88.93Recall: 90.76

논문 -

SICK

Dependency Tree-LSTM (Tai et al., 2015)

MSE: 0.2532Pearson Correlation: 0.8676Spearman Correlation: 0.8083

논문 코드

BIOSSES

BioLinkBERT (large)

Pearson Correlation: 0.9363

논문 코드

CHIP-STS

MacBERT-large

Macro F1: 85.6

논문 코드

ClinicalSTS

CharacterBERT (base, medical, ensemble)

Pearson Correlation: 85.62

논문 코드

MedSTS

NCBI_BERT(base) (P+M)

Pearson Correlation: 0.848

논문 코드

JSTS

row 없음

- -

MNIST

row 없음

- -

Unknown

row 없음

- -

dev evaluation

row 없음

- -

sts dev

row 없음

- -

데이터셋

28개

Annotated corpus for semantic similarity of clinical trial outcomes (expanded corpus)Annotated corpus for semantic similarity of clinical trial outcomes (original corpus)SICK BIOSSES CHIP-STS ClinicalSTS MedSTS JSTS MNIST Unknown dev evaluation sts dev

논문

Measuring semantic similarity of clinical trial outcomes using deep pre-trained language representations

2019-10-17 이 페이지의 평가 5건

F1: 89.75Precision: 88.93Recall: 90.76

Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks

1503.00075 2015-02-28 이 페이지의 평가 3건

MSE: 0.2532Pearson Correlation: 0.8676Spearman Correlation: 0.8083

CharacterBERT: Reconciling ELMo and BERT for Word-Level Open-Vocabulary Representations From Characters

2010.10392 2020-10-20 이 페이지의 평가 1건

Pearson Correlation: 85.62

Efficient Vector Representation for Documents through Corruption

1707.02377 2017-07-08 이 페이지의 평가 1건

MSE: 0.3053Pearson Correlation: 0.8381Spearman Correlation: 0.7621

Skip-Thought Vectors

1506.06726 2015-06-22 이 페이지의 평가 1건

MSE: 0.2687Pearson Correlation: 0.8584Spearman Correlation: 0.7916

Transfer Learning in Biomedical Natural Language Processing: An Evaluation of BERT and ELMo on Ten Benchmarking Datasets

1906.05474 2019-06-13 이 페이지의 평가 1건

Pearson Correlation: 0.848

라이브러리

Papers with Code 리더보드 row에 연결된 구현 저장소입니다.

github.com/dasguptar/treelstm.pytorch 1 논문

github.com/dmlc/dgl/tree/master/examples/mxnet/tree_lstm 1 논문

github.com/dmlc/dgl/tree/master/examples/pytorch/tree_lstm 1 논문

github.com/EmilReinert/DeepLearningPipelines 1 논문

github.com/inyukwo1/tree-lstm 1 논문

github.com/jayanti-prasad/TreeLSTM 1 논문

github.com/Mind23-2/MindCode-17 1 논문

github.com/munashe5/SemanticTreeLSTM 1 논문

Sentence Pair Modeling

벤치마크

데이터셋

논문

Measuring semantic similarity of clinical trial outcomes using deep pre-trained language representations

Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks

CharacterBERT: Reconciling ELMo and BERT for Word-Level Open-Vocabulary Representations From Characters

Efficient Vector Representation for Documents through Corruption

Skip-Thought Vectors

Transfer Learning in Biomedical Natural Language Processing: An Evaluation of BERT and ELMo on Ten Benchmarking Datasets

라이브러리

하위 태스크