Datasets:
ArXiv:
License:
Saptarshi7
commited on
Commit
•
ab084ac
1
Parent(s):
dcc956a
Update README.md
Browse files
README.md
CHANGED
@@ -2,4 +2,4 @@
|
|
2 |
license: apache-2.0
|
3 |
---
|
4 |
|
5 |
-
This is the _cleaned_ version of Deepset's [COVID-QA dataset](https://aclanthology.org/2020.nlpcovid19-acl.18/) which is described in section 3.2 of [
|
|
|
2 |
license: apache-2.0
|
3 |
---
|
4 |
|
5 |
+
This is the _cleaned_ version of Deepset's [COVID-QA dataset](https://aclanthology.org/2020.nlpcovid19-acl.18/) which is described in section 3.2 of [Towards Efficient Methods in Medical Question Answering using Knowledge Graph Embeddings](https://arxiv.org/abs/2401.07977). While you can use either version of the dataset, we recommend using this version since we have corrected many mistakes in the original question set & also, it yields better scores, even zero-shot, as reported in our paper.
|