Sentence Embeddings For Biomedical Texts On

Metrics

Pearson Correlation

Results

Performance results of various models on this benchmark

		Paper Title
Supervised combination of: Jaccard, Q-gram, sent2vec, Paragraph vector DM, skip-thoughts, fastText	0.871	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
Unsupervised combination (mean) of: Jaccard, q-gram, Paragraph vector (PV-DBOW) and sent2vec	0.846	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
Paragraph vector (PV-DM)	0.819	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
BioSentVec (PubMed)	0.817	BioSentVec: creating sentence embeddings for biomedical texts
Paragraph vector (PV-DBOW)	0.804	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
Sent2vec	0.798	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
BioSentVec (PubMed + MIMIC-III)	0.795	BioSentVec: creating sentence embeddings for biomedical texts
Paragraph Vector	0.787	BIOSSES: A Semantic Sentence Similarity Estimation System for the Biomedical Domain
fastText (skip-gram, max pooling)	0.766	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
Q-gram (q = 3)	0.723	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
Skip-thoughts	0.485	Neural sentence embedding models for semantic similarity estimation in the biomedical domain
BioSentVec (MIMIC-III)	0.350	BioSentVec: creating sentence embeddings for biomedical texts
Universal Sentence Encoder	0.345	BioSentVec: creating sentence embeddings for biomedical texts
fastText (CBOW, max pooling)	0.253	Neural sentence embedding models for semantic similarity estimation in the biomedical domain

0 of 14 row(s) selected.

Sentence Embeddings For Biomedical Texts On | SOTA | HyperAI