Search alternatives:
significant decrease » significant increase (Expand Search), significantly increased (Expand Search)
bins decrease » point decrease (Expand Search), sizes decrease (Expand Search), nn decrease (Expand Search)
a decrease » _ decrease (Expand Search), _ decreased (Expand Search), _ decreases (Expand Search)
significant decrease » significant increase (Expand Search), significantly increased (Expand Search)
bins decrease » point decrease (Expand Search), sizes decrease (Expand Search), nn decrease (Expand Search)
a decrease » _ decrease (Expand Search), _ decreased (Expand Search), _ decreases (Expand Search)
-
3861
-
3862
-
3863
-
3864
-
3865
-
3866
-
3867
-
3868
-
3869
-
3870
-
3871
-
3872
-
3873
-
3874
-
3875
S1 File -
Published 2025“…Referring to LastBERT, a customized student BERT model, we significantly lowered model parameters from 110 million BERT base to 29 million-resulting in a model approximately 73.64% smaller. …”
-
3876
Confusion matrix for ClinicalBERT model.
Published 2025“…Referring to LastBERT, a customized student BERT model, we significantly lowered model parameters from 110 million BERT base to 29 million-resulting in a model approximately 73.64% smaller. …”
-
3877
Confusion matrix for LastBERT model.
Published 2025“…Referring to LastBERT, a customized student BERT model, we significantly lowered model parameters from 110 million BERT base to 29 million-resulting in a model approximately 73.64% smaller. …”
-
3878
Student model architecture.
Published 2025“…Referring to LastBERT, a customized student BERT model, we significantly lowered model parameters from 110 million BERT base to 29 million-resulting in a model approximately 73.64% smaller. …”
-
3879
Configuration of the LastBERT model.
Published 2025“…Referring to LastBERT, a customized student BERT model, we significantly lowered model parameters from 110 million BERT base to 29 million-resulting in a model approximately 73.64% smaller. …”
-
3880
Confusion matrix for DistilBERT model.
Published 2025“…Referring to LastBERT, a customized student BERT model, we significantly lowered model parameters from 110 million BERT base to 29 million-resulting in a model approximately 73.64% smaller. …”