diff --git a/sigs/mdp/README.md b/sigs/mdp/README.md
index df60dafcc2e7d7524671a7294a62c738cb4936ed..8b41831a52b5d66be19e8e2de00c7d5b350fdb4f 100644
--- a/sigs/mdp/README.md
+++ b/sigs/mdp/README.md
@@ -2,7 +2,7 @@
 This is the working repo for the MDP Special Interest Group (SIG). MindSpore Deep Probabilistic Programming (MDP) is a programming library for Bayesian deep learning. The target of MDP is to intergrade the gap between deep learning and Bayesian learning. This repo contains all the  artifacts,  materials, meeting notes and proposals regarding  **Probabilistic Programming**  ,  **Deep Probabilistic Programming** ,  **Toolbox** . Feedbacks and contributions are welcomed.
 
 1. Probabilistic Programming: Probabilistic Programming (PP) focuses on professional Bayesian learning, incluiding statistical distributions classes used to generate stochastic tensors and probabilistic inference algorithms.
-2. Deep Probabilistic Programming: Deep Probabilistic Programming (DPP) aims to provide composable BNN modules, which contains bnn layers, bnn, transforms and context.
+2. Deep Probabilistic Programming: Deep Probabilistic Programming (DPP) aims to provide composable BNN modules, which contains bnn layers, bnn modules, transforms and context.
 3. Toolbox: Toolbox provides a set of BNN tools for some specific applications, sunch as Uncertainty Estimation, OoD Detection and so on.
 ## SIG Leads
 Chen Jianfei (Tsinghua University)
@@ -13,4 +13,4 @@ Chen Jianfei (Tsinghua University)
 ## Discussion
 - Slack channel: https://app.slack.com/client/T018BLCMSGL/learning-slack
 - Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/mdp
-## Meeting notes
\ No newline at end of file
+## Meeting notes
diff --git a/sigs/mdp/docs/.keep b/sigs/mdp/docs/.keep
deleted file mode 100644
index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000
diff --git a/sigs/mdp/meetings/.keep b/sigs/mdp/meetings/.keep
deleted file mode 100644
index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000