diff --git a/egs/aishell/ASR/conformer_ctc/transformer.py b/egs/aishell/ASR/conformer_ctc/transformer.py
index dfd888414..a3e50e385 100644
--- a/egs/aishell/ASR/conformer_ctc/transformer.py
+++ b/egs/aishell/ASR/conformer_ctc/transformer.py
@@ -545,7 +545,6 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
-        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 
diff --git a/egs/gigaspeech/ASR/conformer_ctc/transformer.py b/egs/gigaspeech/ASR/conformer_ctc/transformer.py
index 2d797cc67..0566cfc81 100644
--- a/egs/gigaspeech/ASR/conformer_ctc/transformer.py
+++ b/egs/gigaspeech/ASR/conformer_ctc/transformer.py
@@ -549,7 +549,6 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
-        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 
diff --git a/egs/librispeech/ASR/conformer_ctc/transformer.py b/egs/librispeech/ASR/conformer_ctc/transformer.py
index 2d797cc67..0566cfc81 100644
--- a/egs/librispeech/ASR/conformer_ctc/transformer.py
+++ b/egs/librispeech/ASR/conformer_ctc/transformer.py
@@ -549,7 +549,6 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
-        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 
diff --git a/egs/librispeech/ASR/conformer_ctc2/transformer.py b/egs/librispeech/ASR/conformer_ctc2/transformer.py
index 6b62a5993..d3443dc94 100644
--- a/egs/librispeech/ASR/conformer_ctc2/transformer.py
+++ b/egs/librispeech/ASR/conformer_ctc2/transformer.py
@@ -550,7 +550,6 @@ class TransformerDecoderLayer(nn.Module):
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
         warmup: float = 1.0,
-        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 
diff --git a/egs/librispeech/ASR/conformer_mmi/transformer.py b/egs/librispeech/ASR/conformer_mmi/transformer.py
index 3bc6b88ec..2542d9abe 100644
--- a/egs/librispeech/ASR/conformer_mmi/transformer.py
+++ b/egs/librispeech/ASR/conformer_mmi/transformer.py
@@ -537,7 +537,6 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
-        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 
diff --git a/egs/librispeech/ASR/streaming_conformer_ctc/transformer.py b/egs/librispeech/ASR/streaming_conformer_ctc/transformer.py
index 987a45b1f..0c87fdf1b 100644
--- a/egs/librispeech/ASR/streaming_conformer_ctc/transformer.py
+++ b/egs/librispeech/ASR/streaming_conformer_ctc/transformer.py
@@ -567,7 +567,6 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
-        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.