Update README.md

README.md

@@ -6,13 +6,13 @@ tags:
 
 ---
 
-We created a German Question Answering dataset (**deQuAD**) with **130k** training and **11k** test QA pairs and fine-tuned German Question Answering model with [bert-base-multilingual-cased](https://github.com/google-research/bert/blob/master/multilingual.md)  
+We fine-tuned a German Question Answering model based on [electra-base-german-uncased](https://huggingface.co/german-nlp-group/electra-base-german-uncased) with our own German Question Answering dataset (**deQuAD**) containing **130k** training and **11k** test QA pairs.
 
 ## Overview
-- **Language model:** bert-base-multilingual-cased  
-- **Language:** Germa
-- **Training data:** deQuAD2.0 + SQuAD2.0 training set  
-- **Evaluation data:** SQuAD2.0 test set; deQuAD2.0 test set
+- **Language model:** electra-base-german-uncased 
+- **Language:** German
+- **Training data:** deQuAD2.0 training set (~42MB)
+- **Evaluation data:** deQuAD2.0 test set (~4MB)
 - **Infrastructure:** 8xV100 GPU  
 ### Evaluation
 We benchmarked the question answering performance on our deQuAD test data.