diff --git a/mindspore/ops/operations/array_ops.py b/mindspore/ops/operations/array_ops.py
index 94e64c23d04f7835ba21d6301e0888556c9456b3..694c4f1d74b011a7e4df5b7cea8d45871d024104 100644
--- a/mindspore/ops/operations/array_ops.py
+++ b/mindspore/ops/operations/array_ops.py
@@ -2241,7 +2241,8 @@ class ScatterMax(PrimitiveWithInfer):
     """
     Update the value of the input tensor through the max operation.
 
-    Using given values to update tensor value through the max operation, along with the input indices,.
+    Using given values to update tensor value through the max operation, along with the input indices.
+    This operation outputs the `input_x` after the update is done, which makes it convenient to use the updated value.
 
     Args:
         use_locking (bool): Whether protect the assignment by a lock. Default: True.
@@ -2253,7 +2254,7 @@ class ScatterMax(PrimitiveWithInfer):
           the data type is same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
 
     Outputs:
-        Tensor, has the same shape and data type as `input_x`.
+        Parameter, the updated `input_x`.
 
     Examples:
         >>> input_x = Parameter(Tensor(np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]]), mindspore.float32), name="input_x")
@@ -2286,6 +2287,7 @@ class ScatterAdd(PrimitiveWithInfer):
     Update the value of the input tensor through the add operation.
 
     Using given values to update tensor value through the add operation, along with the input indices.
+    This operation outputs the `input_x` after the update is done, which makes it convenient to use the updated value.
 
     Args:
         use_locking (bool): Whether protect the assignment by a lock. Default: False.
@@ -2297,7 +2299,7 @@ class ScatterAdd(PrimitiveWithInfer):
           the data type is same as `input_x`, the shape is `indices_shape + x_shape[1:]`.
 
     Outputs:
-        Tensor, has the same shape and data type as `input_x`.
+        Parameter, the updated `input_x`.
 
     Examples:
         >>> input_x = Parameter(Tensor(np.array([[0.0, 0.0, 0.0], [0.0, 0.0, 0.0]]), mindspore.float32), name="x")
diff --git a/mindspore/ops/operations/nn_ops.py b/mindspore/ops/operations/nn_ops.py
index 0701db8c3bef4a84bf4adf4c62bfb096b54e11f2..dd3339fe8a74d4416b4a202e40b045d136545170 100644
--- a/mindspore/ops/operations/nn_ops.py
+++ b/mindspore/ops/operations/nn_ops.py
@@ -3407,12 +3407,12 @@ class BasicLSTMCell(PrimitiveWithInfer):
     Outputs:
         - **ct** (Tensor) - Forward :math:`c_t` cache at moment `t`. Tensor of shape (`batch_size`, `hidden_size`).
         - **ht** (Tensor) - Cell output. Tensor of shape (`batch_size`, `hidden_size`).
-        - **it** (Tensor) - Forward :math:`i_t` cache at moment `t`. Tensor of shape (`batch_size`, `4 x hidden_size`).
-        - **jt** (Tensor) - Forward :math:`j_t` cache at moment `t`. Tensor of shape (`batch_size`, `4 x hidden_size`).
-        - **ft** (Tensor) - Forward :math:`f_t` cache at moment `t`. Tensor of shape (`batch_size`, `4 x hidden_size`).
-        - **ot** (Tensor) - Forward :math:`o_t` cache at moment `t`. Tensor of shape (`batch_size`, `4 x hidden_size`).
+        - **it** (Tensor) - Forward :math:`i_t` cache at moment `t`. Tensor of shape (`batch_size`, `hidden_size`).
+        - **jt** (Tensor) - Forward :math:`j_t` cache at moment `t`. Tensor of shape (`batch_size`, `hidden_size`).
+        - **ft** (Tensor) - Forward :math:`f_t` cache at moment `t`. Tensor of shape (`batch_size`, `hidden_size`).
+        - **ot** (Tensor) - Forward :math:`o_t` cache at moment `t`. Tensor of shape (`batch_size`, `hidden_size`).
         - **tanhct** (Tensor) - Forward :math:`tanh c_t` cache at moment `t`.
-          Tensor of shape (`batch_size`, `4 x hidden_size`).
+          Tensor of shape (`batch_size`, `hidden_size`).
 
     Examples:
          'block': P.BasicLSTMCell(keep_prob=1.0, forget_bias=1.0, state_is_tuple=True, activation='tanh'),