Update README.md

README.md

@@ -12,13 +12,13 @@ widget:
 ---
 # Model Card for PubMedCLIP
 
-PubMedCLIP is a fine-tuned version of [CLIP](https://huggingface.co/openai/clip-vit-base-patch32) for the medical domain trained on a large number of medical 
+PubMedCLIP is a fine-tuned version of [CLIP](https://huggingface.co/docs/transformers/model_doc/clip) for the medical domain trained on a large number of medical 
 image–text pairs obtained from [PubMed](https://pubmed.ncbi.nlm.nih.gov/) articles.
 
 ## Model Details
 PubMedCLIP was trained on the [Radiology Objects in COntext (ROCO)](https://link.springer.com/chapter/10.1007/978-3-030-01364-6_20) dataset, which provides 
-over 80,000 samples including diverse imaging modalities (e.g. ultrasound, X-Ray, MRI, etc.) from various human body regions (e.g., head, neck, spine, etc.) 
-captured from PubMed articles. 
+over 80,000 samples including diverse imaging modalities (such as ultrasound, X-Ray, MRI, etc.) from various human body regions (such as head, neck, spine, etc.) 
+captured from PubMed articles. The texts used for training were taken from the short captions (average length of 20 words) associated with images in the articles.
 
 ### Model Description
 <!-- Provide a longer summary of what this model is. -->