diff --git a/fluid/PaddleCV/video/README.md b/fluid/PaddleCV/video/README.md
index 1c60974795765b82a5a9df1ad3e08141c8b9f10f..6127ff978c72556d08e4984151d4693a866dbd33 100644
--- a/fluid/PaddleCV/video/README.md
+++ b/fluid/PaddleCV/video/README.md
@@ -13,7 +13,7 @@ Paddle视频模型库第一期主要包含如下模型。
 
 | 模型 | 类别  | 描述 |
 | :---------------: | :--------: | :------------: |
-| [Attention Cluster](./models/attention_cluster/README.md) | 视频分类| CVPR'18提出的视频多模态特征注意力聚簇融合方法 |
+| [Attention Cluster](./models/attention_cluster/README.md) [[论文](https://arxiv.org/abs/1711.09550)] | 视频分类| CVPR'18提出的视频多模态特征注意力聚簇融合方法 |
 | [Attention LSTM](./models/attention_lstm/README.md) | 视频分类| 常用模型，速度快精度高 |
 | [NeXtVLAD](./models/nextvlad/README.md) | 视频分类| 2nd-Youtube-8M最优单模型 |
 | [StNet](./models/stnet/README.md) | 视频分类| AAAI'19提出的视频联合时空建模方法 |
diff --git a/fluid/PaddleCV/video/models/attention_cluster/README.md b/fluid/PaddleCV/video/models/attention_cluster/README.md
index e9f5790954ac492265255424fbcfa39fc4e3d133..64e4c27f60228df8c7dcf89c1cd844f32b1031c3 100644
--- a/fluid/PaddleCV/video/models/attention_cluster/README.md
+++ b/fluid/PaddleCV/video/models/attention_cluster/README.md
@@ -102,8 +102,3 @@ Attention Cluster模型使用2nd-Youtube-8M数据集, 数据下载及准备请
 
 - [Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification](https://arxiv.org/abs/1711.09550), Xiang Long, Chuang Gan, Gerard de Melo, Jiajun Wu, Xiao Liu, Shilei Wen
 
-## 版本更新
-
-- 3/2019: 新增模型
-
-
diff --git a/fluid/PaddleCV/video/models/attention_lstm/README.md b/fluid/PaddleCV/video/models/attention_lstm/README.md
index 4268b73445097b86924bfd396418c1167a760977..7c98ef76e83bc9ea649e7b038393fee38f503696 100644
--- a/fluid/PaddleCV/video/models/attention_lstm/README.md
+++ b/fluid/PaddleCV/video/models/attention_lstm/README.md
@@ -95,7 +95,3 @@ AttentionLSTM模型使用2nd-Youtube-8M数据集，关于数据本分请参考[
 
 - [Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification](https://arxiv.org/abs/1711.09550), Xiang Long, Chuang Gan, Gerard de Melo, Jiajun Wu, Xiao Liu, Shilei Wen
 
-## 版本更新
-
-- 3/2019: 新增模型
-
diff --git a/fluid/PaddleCV/video/models/nextvlad/README.md b/fluid/PaddleCV/video/models/nextvlad/README.md
index 1583f8895d17b911459b1dc8d6070d1d28064ab4..4ba0745db564c0f69d33afa4c907dc27f9a0afbe 100644
--- a/fluid/PaddleCV/video/models/nextvlad/README.md
+++ b/fluid/PaddleCV/video/models/nextvlad/README.md
@@ -87,7 +87,3 @@ NeXtVLAD模型使用2nd-Youtube-8M数据集, 数据下载及准备请参考[数
 
 - [NeXtVLAD: An Efficient Neural Network to Aggregate Frame-level Features for Large-scale Video Classification](https://arxiv.org/abs/1811.05014), Rongcheng Lin, Jing Xiao, Jianping Fan
 
-## 版本更新
-
-- 3/2019: 新增模型
-
diff --git a/fluid/PaddleCV/video/models/stnet/README.md b/fluid/PaddleCV/video/models/stnet/README.md
index 8b654312a37d99ecba0c844dfdb2794af3d63f64..0d7cce67d4eb20b39a2cfa91df2ef55343d97edc 100644
--- a/fluid/PaddleCV/video/models/stnet/README.md
+++ b/fluid/PaddleCV/video/models/stnet/README.md
@@ -106,7 +106,3 @@ StNet的训练数据采用由DeepMind公布的Kinetics-400动作识别数据集
 
 [StNet:Local and Global Spatial-Temporal Modeling for Human Action Recognition](https://arxiv.org/abs/1811.01549), Dongliang He, Zhichao Zhou, Chuang Gan, Fu Li, Xiao Liu, Yandong Li, Limin Wang, Shilei Wen
 
-## 版本更新
-
-- 3/2019: 新增模型
-
diff --git a/fluid/PaddleCV/video/models/tsn/README.md b/fluid/PaddleCV/video/models/tsn/README.md
index 2870d1d847a7f6806a3ea05181e5eb874599362d..52200f9c7b1a4bd3ca32854f4861dea0c2b897cb 100644
--- a/fluid/PaddleCV/video/models/tsn/README.md
+++ b/fluid/PaddleCV/video/models/tsn/README.md
@@ -82,7 +82,3 @@ TSN的训练数据采用由DeepMind公布的Kinetics-400动作识别数据集。
 
 - [StNet:Local and Global Spatial-Temporal Modeling for Human Action Recognition](https://arxiv.org/abs/1608.00859), Limin Wang, Yuanjun Xiong, Zhe Wang, Yu Qiao, Dahua Lin, Xiaoou Tang, Luc Van Gool
 
-## 版本更新
-
-- 3/2019: 新增模型
-