Upload README.md with huggingface_hub

README.md

@@ -63157,7 +63157,7 @@ library_name: sentence-transformers
 
 # BGE-M3 fine-tuned with Matryoshka + MNRLoss
 
-This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) on the json dataset. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
+This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) on the json dataset. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. The main difference with InfectaVec-v1 model is that InfectaVec-v2 is trained with paraphrased and bitext mined queries (En to Kr, Kr to En).
 
 ## Model Details
 
@@ -63203,7 +63203,7 @@ Then you can load this model and run inference.
 from sentence_transformers import SentenceTransformer
 
 # Download from the 🤗 Hub
-model = SentenceTransformer("Yesimm/InfectaVec-v2")
+model = SentenceTransformer("Yesimm/InfectaVec-v1")
 # Run inference
 sentences = [
     '최근 몇 년간 SFTS의 발생 추세는 어떤가요?',
@@ -63422,6 +63422,34 @@ You can finetune this model on your own dataset.
 - Datasets: 4.0.0
 - Tokenizers: 0.21.1
 
+
+### Evaluation Results
+## Evaluation Results on Infectious Diseases Test Dataset
+
+| Model         | Epoch | Accuracy(@1) | Recall(@1) | Precision(@10) | NDCG(@10) | MRR(@10) | MAP(@100) |
+|--------------|-------|-------------|-----------|----------------|-----------|----------|-----------|
+| **BGE-M3**   | -     | 44.49       | 44.49     | 7.43           | 58.97     | 54.12    | 54.83    |
+| **InfectaVec v1** | 2 | 62.67       | 62.67     | 9.37           | 78.21     | 73.23    | 73.58    |
+|              | 3     | 62.30       | 62.30     | 9.42           | 78.58     | 73.52    | 73.85    |
+|              | 4     | 62.83       | 62.83     | 9.46           | 78.92     | 73.87    | 74.18    |
+| **InfectaVec v2** | 2 | 59.49       | 59.49     | 9.08           | 75.35     | 70.37    | 70.81    |
+|              | 3     | 61.08       | 61.08     | 9.16           | 76.43     | 71.55    | 71.98    |
+|              | 4     | 61.29       | 61.29     | 9.16           | 76.49     | 71.63    | 72.07    |
+
+
+## Evaluation Results on MTEB Medical Benchmarks for Retrieval, Clustering and Semantic Text Similarity Tasks
+| Models               | PublicHealthQA (Kr) | PublicHealthQA (En) | MedrxivClusteringS2S.v2 (En) | BIOSSES (En) |
+|---------------------|-------------------|--------------------|-----------------------------|-------------|
+| **BGE-M3**          | 80.41             | 83.81              | 30.63                       | 83.38       |
+| **Multilingual e5-large** | 85.14       | 84.57              | 39.14                       | 87.45       |
+| **InfectaVec-v1**   | 79.70             | 82.57              | 34.62                       | 79.37       |
+| **Qwen-3 Embedding-0.6B** | 81.10       | 83.84              | 40.38                       | 84.73       |
+| **InfectaVec-v2**   | 82.36             | 84.85              | 34.23                       | 76.51       |
+
+## Citation
+
+### BibTeX
+
 ## Citation
 
 ### BibTeX