Skip to Main Content
Table 3: 

Accuracy of binary (mono/poly) and multi-class (poly bands) classifiers using SelfSim and pairCos features on the test sets. Comparison to a baseline that predicts always the same class and a classifier that only uses log frequency as feature. Subscripts denote the layers used.

mono/polypoly bands
ModelSelfSimpairCosSelfSimpairCos
en BERT 0.7610 0.798 0.4910 0.4610 
mBERT 0.778 0.758 0.4612 0.4312 
ELMo 0.692 0.633 0.372 0.343 
context2vec 0.61 0.61 0.34 0.31 
 
Frequency 0.77 0.41 
 
FR Flaubert 0.587 0.556 0.298 0.279 
mBERT 0.669 0.649 0.387 0.388 
 
Frequency 0.61 0.37 
 
ES BETO 0.709 0.667 0.426 0.485 
mBERT 0.6911 0.647 0.389 0.437 
 
Frequency 0.67 0.41 
 
el GreekBERT 0.704 0.644 0.344 0.386 
mBERT 0.607 0.657 0.3211 0.349 
 
Frequency 0.63 0.35 
 
 Baseline 0.50 0.25 
mono/polypoly bands
ModelSelfSimpairCosSelfSimpairCos
en BERT 0.7610 0.798 0.4910 0.4610 
mBERT 0.778 0.758 0.4612 0.4312 
ELMo 0.692 0.633 0.372 0.343 
context2vec 0.61 0.61 0.34 0.31 
 
Frequency 0.77 0.41 
 
FR Flaubert 0.587 0.556 0.298 0.279 
mBERT 0.669 0.649 0.387 0.388 
 
Frequency 0.61 0.37 
 
ES BETO 0.709 0.667 0.426 0.485 
mBERT 0.6911 0.647 0.389 0.437 
 
Frequency 0.67 0.41 
 
el GreekBERT 0.704 0.644 0.344 0.386 
mBERT 0.607 0.657 0.3211 0.349 
 
Frequency 0.63 0.35 
 
 Baseline 0.50 0.25 
Close Modal

or Create an Account

Close Modal
Close Modal