diff --git a/official/nlp/mass/README.md b/official/nlp/mass/README.md
index 0df1841d93ad018c89d56925dcf8c825749ec34f..eb3826d455c44d9c652ca9e5028a85921e0cf0fd 100644
--- a/official/nlp/mass/README.md
+++ b/official/nlp/mass/README.md
@@ -69,11 +69,9 @@ Note that you can run the scripts based on the dataset mentioned in original pap
 
 Dataset used:
 
-- monolingual English data from News Crawl dataset(WMT 2019) for pre-training.
-- Gigaword Corpus(Graff et al., 2003) for Text Summarization.
-- Cornell movie dialog corpus(DanescuNiculescu-Mizil & Lee, 2011).
-
-Details about those dataset could be found in [MASS: Masked Sequence to Sequence Pre-training for Language Generation](https://www.microsoft.com/en-us/research/uploads/prod/2019/06/MASS-paper-updated-002.pdf).
+- [monolingual English data from News Crawl dataset](https://www.statmt.org/wmt16/translation-task.html)(WMT 2019) for pre-training.
+- [Gigaword Corpus](https://github.com/harvardnlp/sent-summary)(Graff et al., 2003) for Text Summarization.
+- [Cornell movie dialog corpus](https://github.com/suriyadeepan/datasets/tree/master/seq2seq/)(DanescuNiculescu-Mizil & Lee, 2011).
 
 # Features
 
diff --git a/official/nlp/mass/README_CN.md b/official/nlp/mass/README_CN.md
index fd0503339e70026473c893930d97b27b8e84f536..65b47f13c3c7495b9f9b8d475d6d59e93d15b5dc 100644
--- a/official/nlp/mass/README_CN.md
+++ b/official/nlp/mass/README_CN.md
@@ -67,11 +67,9 @@ MASS网络由Transformer实现，Transformer包括多个编码器层和多个解
 
 本文运用数据集包括：
 
-- News Crawl数据集（WMT，2019年）的英语单语数据，用于预训练
-- Gigaword语料库（Graff等人，2003年），用于文本摘要
-- Cornell电影对白语料库（DanescuNiculescu-Mizil & Lee，2011年）
-
-数据集相关信息，参见[MASS：语言生成的隐式序列到序列预训练](https://www.microsoft.com/en-us/research/uploads/prod/2019/06/MASS-paper-updated-002.pdf)。
+- [News Crawl数据集](https://www.statmt.org/wmt16/translation-task.html)（WMT，2019年）的英语单语数据，用于预训练
+- [Gigaword语料库](https://github.com/harvardnlp/sent-summary)（Graff等人，2003年），用于文本摘要
+- [Cornell电影对白语料库](https://github.com/suriyadeepan/datasets/tree/master/seq2seq/)（DanescuNiculescu-Mizil & Lee，2011年）
 
 ## 特性