Update README.md

README.md

@@ -1,3 +1,17 @@
----
-license: cc-by-2.5
----
+---
+license: cc-by-2.5
+language:
+- nl
+tags:
+- medical
+pretty_name: BioASQ Dutch, Q/A
+size_categories:
+- 1K<n<10K
+task_categories:
+- question-answering
+- sentence-similarity
+---
+
+[Bio ASQ challenge 11b](https://bioasq.org/)
+
+This can be used to finetune a decoder model for Q/A interaction, alternatively it can be used to create (question, positive, negative) triplets to train a sentence encoder using [SBERT](https://www.sbert.net/).