Natural Language Inference on SNLI

Metric: Dev Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	Dev Accuracy▼	Extra Data	Paper	Date↕	Code
1	MT-DNN-SMART_100%ofTrainingData	91.6	No	SMART: Robust and Efficient Fine-Tuning for Pre-...	2019-11-08	Code
2	MT-DNN-SMART_10%ofTrainingData	88.7	No	SMART: Robust and Efficient Fine-Tuning for Pre-...	2019-11-08	Code
3	MT-DNN-SMART_1%ofTrainingData	86	No	SMART: Robust and Efficient Fine-Tuning for Pre-...	2019-11-08	Code
4	MT-DNN-SMART_0.1%ofTrainingData	82.7	No	SMART: Robust and Efficient Fine-Tuning for Pre-...	2019-11-08	Code

#1MT-DNN-SMART_100%ofTrainingDataSOTA
91.6
Dev Accuracy· 2019-11-08
SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization Code
#2MT-DNN-SMART_10%ofTrainingData
88.7
Dev Accuracy· 2019-11-08
SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization Code
#3MT-DNN-SMART_1%ofTrainingData
86
Dev Accuracy· 2019-11-08
SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization Code
#4MT-DNN-SMART_0.1%ofTrainingData
82.7
Dev Accuracy· 2019-11-08
SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization Code