Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Penn Treebank

Benchmarks

Task NameDataset NameSOTA ResultTrend
Language ModelingPenn Treebank (test)
Perplexity46.34
411
Language ModelingPenn Treebank (val)
Perplexity46.64
178
Language ModelingPenn Treebank (PTB) (test)
Perplexity14.72
120
Character-level Language ModelingPenn Treebank (test)
BPC1.175
113
Dependency ParsingPenn Treebank (PTB) (test)
LAS96.4
80
Language ModelingPenn Treebank word-level (test)
Perplexity49.95
72
Language ModelingPenn Treebank (PTB) (val)
Perplexity46.63
70
Part-of-Speech TaggingPenn TreeBank (test)
Accuracy97.96
64
Constituency ParsingPenn Treebank WSJ (section 23 test)
F1 Score95.8
55
Character-level Language ModelingPenn Treebank char-level (test)
BPC1.16
25
Unlabeled ParsingPenn Treebank WSJ (test)
F1 (mean)84.3
25
Dependency ParsingPenn Treebank (PTB) Section 23 v2.2 (test)
UAS95.66
17
Unsupervised Constituency ParsingPenn TreeBank English (test)
Mean S-F169.6
16
Unsupervised ParsingPenn Treebank WSJ Section 23 (test)
F1 Score57.22
15
POS TaggingPenn Treebank (PTB) Section 23 v2.2 (test)
POS Accuracy97.97
15
Constituency ParsingPenn Treebank shortest 25% of samples < 128 tokens (test)
Bracket Precision76.6
14
Language ModelingPenn Treebank (dev)
Perplexity (PPL)56.5
14
Unlabeled ParsingPenn Treebank WSJ10 (test)
F1 (max)82.9
14
Syntactic ParsingEnglish Penn Treebank (test)
Speed (Sents/s)1,127
11
Language ModelingPenn Treebank (PTB) word-level (val)
Perplexity56.5
11
Word OrderingPenn Treebank (test)
BLEU34.5
11
Part-of-speech taggingPenn Treebank POS (test)
F1 Score97.58
10
Character-level language modelingPenn Treebank character-level (val)
BPC1.24
10
Constituency ParsingPenn Treebank WSJ section 22 (dev)
F1 Score93.5
9
5-way few-shot classificationPenn Treebank v1 (test)
1-shot Accuracy (random)72.8
8
Showing 25 of 35 rows