Update README.md

README.md

@@ -16,6 +16,11 @@ We utilize 4 Chinese dialogue datasets from [LUGE](https://www.luge.ai/#/)
 |                              |            |                       |
 
 
+# Data format
+Input: `[CLS] 对话历史：<history> 知识：<knowledge> [SEP]`
+Output: `[CLS] <response> [SEP]`
+
+
 # Example
 ```python
 from transformers import BertTokenizer, BartForConditionalGeneration