Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Natural Language Inference on MNLI GLUE matched mismatched average (test)
Loading...
84.8
Accuracy
TR-BERT
65.248
70.324
75.4
80.476
Mar 16, 2022
May 27, 2022
Aug 8, 2022
Oct 20, 2022
Jan 1, 2023
Mar 15, 2023
May 27, 2023
Accuracy
Speedup (x)
Mod Score
Similarity Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
Speedup (x)
Mod Score
Similarity Score
TR-BERT
Backbone=BERT-base-unc...
2022.03
84.8
1
-
-
BERT
Backbone=BERT-base-unc...
2022.03
84.2
1
-
-
PoWER-BERT
Backbone=BERT-base-unc...
2022.03
82.9
1.1
-
-
AdapLeR
Backbone=BERT-base-unc...
2022.03
82.9
1.42
-
-
DistilBERT
Backbone=BERT-base-unc...
2022.03
81.8
2
-
-
SDM-ATTACK
Victim Model=BERT
2023.05
78.7
-
13.8
49
BERT-Attack
Victim Model=BERT
2023.05
78.1
-
14
46
TextFooler
Victim Model=BERT
2023.05
76.5
-
15
45
A2T
Victim Model=BERT
2023.05
66
-
14.4
45
Feedback
Search any
task
Search any
task