List your company

Semantic Textual Similarity on BIOSSES

89.6Spearman CorrelationBioHiCL-base
37.13250.753564.37577.9965Jun 5, 2024Sep 26, 2024Jan 18, 2025May 11, 2025Sep 2, 2025Dec 24, 2025Apr 17, 2026

Evaluation results

MethodLinks
BioHiCL-baseSize=0.1B2026.0489.6
BiCA-baseSize=0.1B2026.0488
BiCA-smallSize=0.03B2026.0487.2
BioLORDType=Black-Box2026.0387.18
MedEmbedType=Black-Box2026.0386.99
BioHiCL-largeSize=0.3B2026.0486.8
bge-base-en-v1.5Size=0.1B2026.0486
BMRetriever-1BSize=1B2026.0485.8
bge-large-en-v1.5Size=0.3B2026.0484.4
BMRetriever-410MSize=0.4B2026.0484
PubMedBERTType=Black-Box2026.0383.96
MedCPT-QuerySize=0.1B2026.0483.7
ContrieverSize=0.1B2026.0483.3
e5-large-v2Size=0.3B2026.0483
SGPT-1.3BSize=1.3B2026.0483
SpladeV3Size=0.3B2026.0482.7
SimCSE + MixSPEncoder Architecture=BERT-Base, Base Method=SimCSE, Training Strategy=MixSP2024.0682.61
SapBERTType=Black-Box2026.0382.48
MedCPTType=Black-Box2026.0381.95
DiffAug + MixSPEncoder Architecture=BERT-Base, Base Method=DiffAug, Training Strategy=MixSP2024.0681.23
SimCSE + MixSPEncoder Architecture=RoBERTa-Base, Base Method=SimCSE, Training Strategy=MixSP2024.0680.74
SBERT + MixSPEncoder Architecture=BERT-Base, Base Method=SBERT, Training Strategy=MixSP2024.0680.58
EmbeddingGemmaType=Black-Box2026.0380.46
DiffAug + MixSPEncoder Architecture=RoBERTa-Base, Base Method=DiffAug, Training Strategy=MixSP2024.0680.35
DiffAug + MoEEncoder Architecture=BERT-Base, Base Method=DiffAug, Training Strategy=MoE2024.0679.95
QIME-TF-MMRType=Interpretable2026.0379.66
LDIR-500Type=Interpretable2026.0379.3
BiomedBERTSize=0.1B2026.0479.1
gtr-t5-xlSize=1.2B2026.0478.9
SBERT + MoEEncoder Architecture=BERT-Base, Base Method=SBERT, Training Strategy=MoE2024.0678.52
SimCSE + MoEEncoder Architecture=BERT-Base, Base Method=SimCSE, Training Strategy=MoE2024.0677.07
SimCSE + FTEncoder Architecture=BERT-Base, Base Method=SimCSE, Training Strategy=FT2024.0676.63
SBERT + MixSPEncoder Architecture=RoBERTa-Base, Base Method=SBERT, Training Strategy=MixSP2024.0676.01
QIME-TFType=Interpretable2026.0375.6
SBERT + MoEEncoder Architecture=RoBERTa-Base, Base Method=SBERT, Training Strategy=MoE2024.0673.75
SimCSE + FTEncoder Architecture=RoBERTa-Base, Base Method=SimCSE, Training Strategy=FT2024.0673.35
SimCSE + MoEEncoder Architecture=RoBERTa-Base, Base Method=SimCSE, Training Strategy=MoE2024.0672.65
DiffAug + MoEEncoder Architecture=RoBERTa-Base, Base Method=DiffAug, Training Strategy=MoE2024.0672.65
SBERTEncoder Architecture=RoBERTa-Base, Base Method=SBERT, Training Strategy=Base2024.0672.03
DiffAug + FTEncoder Architecture=BERT-Base, Base Method=DiffAug, Training Strategy=FT2024.0671.26
DiffAug + FTEncoder Architecture=RoBERTa-Base, Base Method=DiffAug, Training Strategy=FT2024.0671.02
SimCSE (Unsup)Type=Black-Box2026.0368.86
BMRetrieverType=Black-Box2026.0368.85
Bag-of-WordsType=Interpretable2026.0368.78
SimCSEEncoder Architecture=BERT-Base, Base Method=SimCSE, Training Strategy=Base2024.0668.38
SimCSEEncoder Architecture=RoBERTa-Base, Base Method=SimCSE, Training Strategy=Base2024.0667.75
SimCSE (Sup)Type=Black-Box2026.0367.19
SBERTEncoder Architecture=BERT-Base, Base Method=SBERT, Training Strategy=Base2024.0663.88
QIMEType=Interpretable2026.0361.88
CQG-MBQAType=Interpretable2026.0354.97
BERTType=Black-Box2026.0354.7
QA-EmbType=Interpretable2026.0346.43
GloVeType=Black-Box2026.0344.93
DiffAugEncoder Architecture=BERT-Base, Base Method=DiffAug, Training Strategy=Base2024.0640.12
DiffAugEncoder Architecture=RoBERTa-Base, Base Method=DiffAug, Training Strategy=Base2024.0639.15