BioBERT base
According to the model card, the base model is monologg/biobert_v1.1_pubmed, a biomedical language representation model.
Open Source Model Profile · gsarti
biobert-nli
biobert-nli is a BERT-based feature-extraction model from gsarti. According to the model card, it adapts BioBERT with sentence-transformers on SNLI and MultiNLI to produce universal sentence embeddings for biomedical text.
- Publisher
- gsarti
- Task
- feature-extraction
- Model type
- bert
- License
- Unknown
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
biobert-nli is published by gsarti as a BERT feature-extraction model. The captured configuration identifies BertModel with model type bert and the Transformers library. According to the model card, it starts from monologg/biobert_v1.1_pubmed and is trained with sentence-transformers on SNLI and MultiNLI using average pooling and a softmax loss.
Recorded capabilities
NLI sentence-embedding training
According to the model card, the model was trained on SNLI and MultiNLI with sentence-transformers to produce universal sentence embeddings.
Average pooling with softmax loss
According to the model card, it uses the original BERT wordpiece vocabulary with average pooling and a softmax loss.
Use cases in the source record
- Similarity-based scientific paper retrieval, which the model card illustrates with the Covid Papers Browser repository.
- Biomedical sentence-embedding workflows such as semantic search and clustering built on universal sentence representations.
Limitations and unknowns
- No parameter count was extracted from this record.
- No license value was extracted from this record.
- The publisher-reported STS figure of 77.12 is a source claim and was not independently verified.
- Provider state is historical snapshot data, not independently refreshed current availability.
Source and provenance
Source: gsarti/biobert-nli
Captured: Unknown. Processed: 2026-09-07T19:34:46.001979+00:00.
BioBERT-NLI This is the model BioBERT [1] fine-tuned on the SNLI and the MultiNLI datasets using the sentence-transformers library to produce universal sentence embeddings [2]. The model uses the original BERT wordpiece vocabulary and was trained using the average pooling strategy and a softmax loss . Base model : monologg/biobert_v1.1_pubmed from HuggingFace's AutoModel . Training time : ~6 hours on the NVIDIA Tesla P100 GPU provided in Kaggle Notebooks. Parameters : Parameter Value Batch size 64 Training steps 30000 Warmup steps 1450 Lowercasing False Max. Seq. Length 128 Performances : The performance was evaluated on the test po…
F001F002F003F004F005F006F007F008F009F010F011F012F014F015F016