Natural Language Inference on MultiNLI Dev

Metric: Matched (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	Matched▼	Extra Data	Paper	Date↕	Code
1	TinyBERT-6 67M	84.5	No	TinyBERT: Distilling BERT for Natural Language U...	2019-09-23	Code
2	BERT-Large-uncased-PruneOFA (90% unstruct sparse)	83.74	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
3	BERT-Large-uncased-PruneOFA (90% unstruct sparse, QAT Int8)	83.47	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
4	BERT-Base-uncased-PruneOFA (85% unstruct sparse)	82.71	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
5	BERT-Base-uncased-PruneOFA (90% unstruct sparse)	81.45	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
6	BERT-Base-uncased-PruneOFA (85% unstruct sparse, QAT Int8)	81.4	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
7	DistilBERT-uncased-PruneOFA (85% unstruct sparse)	81.35	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
8	DistilBERT-uncased-PruneOFA (90% unstruct sparse)	80.68	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
9	DistilBERT-uncased-PruneOFA (85% unstruct sparse, QAT Int8)	80.66	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code
10	DistilBERT-uncased-PruneOFA (90% unstruct sparse, QAT Int8)	78.8	No	Prune Once for All: Sparse Pre-Trained Language ...	2021-11-10	Code

#1TinyBERT-6 67MSOTA
84.5
Matched· 2019-09-23
TinyBERT: Distilling BERT for Natural Language Understanding Code
#2BERT-Large-uncased-PruneOFA (90% unstruct sparse)
83.74
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#3BERT-Large-uncased-PruneOFA (90% unstruct sparse, QAT Int8)
83.47
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#4BERT-Base-uncased-PruneOFA (85% unstruct sparse)
82.71
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#5BERT-Base-uncased-PruneOFA (90% unstruct sparse)
81.45
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#6BERT-Base-uncased-PruneOFA (85% unstruct sparse, QAT Int8)
81.4
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#7DistilBERT-uncased-PruneOFA (85% unstruct sparse)
81.35
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#8DistilBERT-uncased-PruneOFA (90% unstruct sparse)
80.68
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#9DistilBERT-uncased-PruneOFA (85% unstruct sparse, QAT Int8)
80.66
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code
#10DistilBERT-uncased-PruneOFA (90% unstruct sparse, QAT Int8)
78.8
Matched· 2021-11-10
Prune Once for All: Sparse Pre-Trained Language Models Code