diff --git a/paddle/operators/shrink_rnn_memory_op.cc b/paddle/operators/shrink_rnn_memory_op.cc
index 48597c1d2ace9cb5fe36ba237f70cab8b280a836..c380e606869fd2c559c7d5f378857ca74fa8d8d3 100644
--- a/paddle/operators/shrink_rnn_memory_op.cc
+++ b/paddle/operators/shrink_rnn_memory_op.cc
@@ -57,11 +57,21 @@ class ShrinkRNNMemoryOpProtoMaker : public framework::OpProtoAndCheckerMaker {
   ShrinkRNNMemoryOpProtoMaker(framework::OpProto *proto,
                               framework::OpAttrChecker *op_checker)
       : OpProtoAndCheckerMaker(proto, op_checker) {
-    AddInput("X", "");
-    AddInput("RankTable", "");
-    AddInput("I", "");
-    AddOutput("Out", "");
-    AddComment("");
+    AddInput("X", "(LoDTensor) The RNN step memory to be shrinked.");
+    AddInput("RankTable", "(LoDRankTable) The lod_rank_table of dynamic RNN.");
+    AddInput("I",
+             "(LoDTensor) The step index. The RNN step memory 'X' will be "
+             "shrinked to match the size of the input of the index'th step.");
+    AddOutput("Out", "(LoDTensor) The shrinked RNN step memory.");
+    AddComment(
+        R"DOC(
+        In dynamic RNN, we are able to handle sequences of different lengths. 
+        Because of the multiple lengths, the size of each step input can be 
+        different, which may lead to a mismatching between the input of
+        the current step and the memory generated by the previous one. This 
+        operator shrinks memory according to the size of the next step input, 
+        to make sure that they can match each other.
+        )DOC");
   }
 };