diff --git a/paddle/gserver/layers/FactorizationMachineLayer.h b/paddle/gserver/layers/FactorizationMachineLayer.h
index 85ac175657c35fd5a266cc914f788d624d4a591e..3bc36daaab3317b7b3a71b8cd417a8ce74f86bc9 100644
--- a/paddle/gserver/layers/FactorizationMachineLayer.h
+++ b/paddle/gserver/layers/FactorizationMachineLayer.h
@@ -36,8 +36,7 @@ namespace paddle {
  *
  * The detailed calculation for forward and backward can be found at this paper:
  *
- *     Rendle, Steffen. Factorization machines. IEEE 10th International
- *     Conference on Data Mining (ICDM). IEEE, 2010.
+ *     Factorization machines.
  *
  * The config file api is factorization_machine.
  */
@@ -59,7 +58,7 @@ private:
   // The result of input matrix * latent vector matrix that will be used in
   // both forward and backward step
   MatrixPtr inputMulFactor_;
-  // Temporary calculation result store
+  // Store temporary calculation result
   MatrixPtr tmpOut_;
   MatrixPtr tmpSum_;
   // Negative identity matrix
diff --git a/python/paddle/trainer_config_helpers/layers.py b/python/paddle/trainer_config_helpers/layers.py
index cc1bf923dd00c51c9d3efed5daca41eb67a5ca07..37214a53d362bdc28884f77cbb020b8c29c0816a 100644
--- a/python/paddle/trainer_config_helpers/layers.py
+++ b/python/paddle/trainer_config_helpers/layers.py
@@ -3876,7 +3876,7 @@ def recurrent_layer(input,
     :type input: LayerOutput
     :param act: Activation type. TanhActivation is the default activation.
     :type act: BaseActivation
-    :param bias_attr: The parameter attribute for bias. If this parameter is set to 
+    :param bias_attr: The parameter attribute for bias. If this parameter is set to
                       False or an object whose type is not ParameterAttribute,
                       no bias is defined. If the parameter is set to True,
                       the bias is initialized to zero.
@@ -7307,8 +7307,7 @@ def factorization_machine(input,
         each latent vector is k.
 
     For details of Factorization Machine, please refer to the paper:
-        Rendle, Steffen. Factorization machines. IEEE 10th International
-        Conference on Data Mining (ICDM). IEEE, 2010.
+    Factorization machines.
 
     .. code-block:: python
        factor_machine = factorization_machine(input=input_layer, factor_size=10)