diff --git a/example/bert_clue/README.md b/example/bert_clue/README.md
index 3c66816ff34adc8b67e2891b3f730e9483e1a220..2e778c968c6fc5276decb03ebda2d241d3953ece 100644
--- a/example/bert_clue/README.md
+++ b/example/bert_clue/README.md
@@ -4,9 +4,8 @@ This example implements pre-training, fine-tuning and evaluation of [BERT-base](
 
 ## Requirements
 - Install [MindSpore](https://www.mindspore.cn/install/en).
-- Download the zhwiki dataset from <https://dumps.wikimedia.org/zhwiki> for pre-training. Extract and clean text in the dataset with [WikiExtractor](https://github.com/attardi/wil
-kiextractor). Convert the dataset to TFRecord format and move the files to a specified path.
-- Download the CLUE dataset from <https://www.cluebenchmarks.com> for fine-tuning and evaluation.
+- Download the zhwiki dataset for pre-training. Extract and clean text in the dataset with [WikiExtractor](https://github.com/attardi/wikiextractor). Convert the dataset to TFRecord format and move the files to a specified path.
+- Download the CLUE dataset for fine-tuning and evaluation.
 >  Notes:
    If you are running a fine-tuning or evaluation task, prepare the corresponding checkpoint file.