Paraphrase Identification On Quora Question

评估指标

Accuracy

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
data2vec	92.4	data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language
Charformer-Tall	91.4	Charformer: Fast Character Transformers via Gradient-based Subword Tokenization
RealFormer	91.34	RealFormer: Transformer Likes Residual Attention
StructBERTRoBERTa ensemble	90.7	StructBERT: Incorporating Language Structures into Pre-training for Deep Language Understanding	-
MFAE	90.54	What Do Questions Exactly Ask? MFAE: Duplicate Question Identification with Multi-Fusion Asking Emphasis	-
XLNet-Large (ensemble)	90.3	XLNet: Generalized Autoregressive Pretraining for Language Understanding
Snorkel MeTaL(ensemble)	89.9	Training Complex Models with Multi-Task Weak Supervision
MT-DNN	89.6	Multi-Task Deep Neural Networks for Natural Language Understanding
SpanBERT	89.5	SpanBERT: Improving Pre-training by Representing and Predicting Spans
RE2	89.2	Simple and Effective Text Matching with Richer Alignment Features
MwAN	89.12	Multiway Attention Networks for Modeling Sentence Pairs	-
DIIN	89.06	Natural Language Inference over Interaction Space
MSEM	88.86	Multi-task Sentence Encoding Model for Semantic Retrieval in Question Answering Systems	-
Bi-CAS-LSTM	88.6	Cell-aware Stacked LSTMs for Modeling Sentences	-
pt-DecAtt	88.40	Neural Paraphrase Identification of Questions with Noisy Pretraining	-
TRANS-BLSTM	88.28	TRANS-BLSTM: Transformer with Bidirectional LSTM for Language Understanding	-
BiMPM	88.17	Bilateral Multi-Perspective Matching for Natural Language Sentences
GenSen	87.01	Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning
1-3[0.8pt/2pt] Random	80	Self-Explaining Structures Improve NLP Models
FreeLB	74.8	SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization

0 of 31 row(s) selected.