diff --git a/PaddleNLP/pretrain_langauge_models/BERT/utils/args.py b/PaddleNLP/pretrain_langauge_models/BERT/utils/args.py
index 0f6700a2fca5217b2f82824d774e91ad385a9144..fb060d46a9cef41a7a28b5169d330f1eac2038b8 100644
--- a/PaddleNLP/pretrain_langauge_models/BERT/utils/args.py
+++ b/PaddleNLP/pretrain_langauge_models/BERT/utils/args.py
@@ -17,6 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
+import sys
 import six
 import argparse
 
diff --git a/PaddleNLP/pretrain_langauge_models/XLNet/utils/args.py b/PaddleNLP/pretrain_langauge_models/XLNet/utils/args.py
index 66e9bb81a35bb4cc4c8c79cac4631841742bdeb8..e11d635bff1d18a8ff8db5a0b4f1f35672f5c11e 100644
--- a/PaddleNLP/pretrain_langauge_models/XLNet/utils/args.py
+++ b/PaddleNLP/pretrain_langauge_models/XLNet/utils/args.py
@@ -17,6 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
+import sys
 import six
 import argparse