Sentence Embeddings on BIOSSES

Metric: Pearson Correlation (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Pearson Correlation▼	Extra Data	Paper	Date↕	Code
1	Supervised combination of: Jaccard, Q-gram, sent2vec, Paragraph vector DM, skip-thoughts, fastText	0.871	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
2	Unsupervised combination (mean) of: Jaccard, q-gram, Paragraph vector (PV-DBOW) and sent2vec	0.846	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
3	Paragraph vector (PV-DM)	0.819	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
4	BioSentVec (PubMed)	0.817	Yes	BioSentVec: creating sentence embeddings for bio...	2018-10-22	Code
5	Paragraph vector (PV-DBOW)	0.804	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
6	Sent2vec	0.798	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
7	BioSentVec (PubMed + MIMIC-III)	0.795	Yes	BioSentVec: creating sentence embeddings for bio...	2018-10-22	Code
8	Paragraph Vector	0.787	Yes	-	-	-
9	fastText (skip-gram, max pooling)	0.766	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
10	Q-gram (q = 3)	0.723	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
11	Skip-thoughts	0.485	No	Neural sentence embedding models for semantic si...	2021-10-01	Code
12	BioSentVec (MIMIC-III)	0.35	Yes	BioSentVec: creating sentence embeddings for bio...	2018-10-22	Code
13	Universal Sentence Encoder	0.345	No	BioSentVec: creating sentence embeddings for bio...	2018-10-22	Code
14	fastText (CBOW, max pooling)	0.253	No	Neural sentence embedding models for semantic si...	2021-10-01	Code

#1Supervised combination of: Jaccard, Q-gram, sent2vec, Paragraph vector DM, skip-thoughts, fastTextSOTA
0.871
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#2Unsupervised combination (mean) of: Jaccard, q-gram, Paragraph vector (PV-DBOW) and sent2vec
0.846
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#3Paragraph vector (PV-DM)
0.819
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#4BioSentVec (PubMed)SOTA
0.817
Pearson Correlation· Extra Data· 2018-10-22
BioSentVec: creating sentence embeddings for biomedical texts Code
#5Paragraph vector (PV-DBOW)
0.804
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#6Sent2vec
0.798
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#7BioSentVec (PubMed + MIMIC-III)
0.795
Pearson Correlation· Extra Data· 2018-10-22
BioSentVec: creating sentence embeddings for biomedical texts Code
#8Paragraph Vector
0.787
Pearson Correlation· Extra Data
No paper
#9fastText (skip-gram, max pooling)
0.766
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#10Q-gram (q = 3)
0.723
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#11Skip-thoughts
0.485
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code
#12BioSentVec (MIMIC-III)
0.35
Pearson Correlation· Extra Data· 2018-10-22
BioSentVec: creating sentence embeddings for biomedical texts Code
#13Universal Sentence Encoder
0.345
Pearson Correlation· 2018-10-22
BioSentVec: creating sentence embeddings for biomedical texts Code
#14fastText (CBOW, max pooling)
0.253
Pearson Correlation· 2021-10-01
Neural sentence embedding models for semantic similarity estimation in the biomedical domain Code