Natural Language Inference On Snli

Métriques

% Test Accuracy

% Train Accuracy

Parameters

Résultats

Résultats de performance de divers modèles sur ce benchmark

				Paper Title
UnitedSynT5 (3B)	94.7	-	-	First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI
UnitedSynT5 (335M)	93.5	-	-	First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI
Neural Tree Indexers for Text Understanding	93.1	-	355	Entailment as Few-Shot Learner
EFL (Entailment as Few-shot Learner) + RoBERTa-large	93.1	?	355m	Entailment as Few-Shot Learner
RoBERTa-large + self-explaining layer	92.3	?	355m+	Self-Explaining Structures Improve NLP Models
RoBERTa-large+Self-Explaining	92.3	-	340	Self-Explaining Structures Improve NLP Models
CA-MTL	92.1	92.6	340m	Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data
SemBERT	91.9	94.4	339m	Semantics-aware BERT for Language Understanding
MT-DNN-SMARTLARGEv0	91.7	-	-	SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization
MT-DNN	91.6	97.2	330m	Multi-Task Deep Neural Networks for Natural Language Understanding
SJRC (BERT-Large +SRL)	91.3	95.7	308m	Explicit Contextual Semantics for Text Comprehension
Ntumpha	90.5	99.1	220	Multi-Task Deep Neural Networks for Natural Language Understanding
Densely-Connected Recurrent and Co-Attentive Network Ensemble	90.1	95.0	53.3m	Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information
MFAE	90.07	93.18	-	What Do Questions Exactly Ask? MFAE: Duplicate Question Identification with Multi-Fusion Asking Emphasis
Fine-Tuned LM-Pretrained Transformer	89.9	96.6	85m	Improving Language Understanding by Generative Pre-Training
300D DMAN Ensemble	89.6	96.1	79m	-
300D DMAN Ensemble	89.6	96.1	79m	Discourse Marker Augmented Network with Reinforcement Learning for Natural Language Inference
150D Multiway Attention Network Ensemble	89.4	95.5	58m	Multiway Attention Networks for Modeling Sentence Pairs
ESIM + ELMo Ensemble	89.3	92.1	40m	Deep contextualized word representations
450D DR-BiLSTM Ensemble	89.3	94.8	45m	DR-BiLSTM: Dependent Reading Bidirectional LSTM for Natural Language Inference

0 of 98 row(s) selected.

Command Palette

Natural Language Inference On Snli

Métriques

Résultats