Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Semantic Textual Similarity on STS-16
Loading...
77.63
Spearman Rho (x100)
BERTlarge-flow (target)
36.446
47.138
57.83
68.522
Nov 2, 2020
Oct 6, 2021
Sep 10, 2022
Aug 15, 2023
Jul 19, 2024
Jun 23, 2025
May 28, 2026
Spearman Rho (x100)
Updated 5d ago
Evaluation Results
Method
Method
Links
Spearman Rho (x100)
BERTlarge-flow (target)
Backbone=BERT-large, P...
2020.11
77.63
BERTbase-flow (target)
Backbone=BERT-base, Po...
2020.11
75.37
BERTlarge-flow (NLI*)
Backbone=BERT-large, P...
2020.11
74.47
MIC
Backbone=BERT, Dimensi...
2026.05
72.18
BERTbase-flow (NLI*)
Backbone=BERT-base, Po...
2020.11
71.84
MIPIC
Backbone=BERT, Dimensi...
2026.04
71.56
BERTlarge-last2avg
Backbone=BERT-large, P...
2020.11
70.32
BERTbase-last2avg
Backbone=BERT-base, Po...
2020.11
69.81
MIC
Backbone=TinyBERT 6L,...
2026.05
69.47
BERTbase
Backbone=BERT-base, Po...
2020.11
65.19
MIC
Backbone=BERT, Dimensi...
2026.05
63.76
Avg. GloVe embeddings
Backbone=GloVe, Poolin...
2020.11
63.66
BERTlarge
Backbone=BERT-large, P...
2020.11
61.63
Avg. BERT embeddings
Backbone=BERT, Pooling...
2020.11
61.06
MIC
Backbone=TinyBERT 6L,...
2026.05
60.64
BERT CLS-vector
Backbone=BERT, Pooling...
2020.11
38.03
Feedback
Search any
task
Search any
task