[docs] Fix docstring arguments that don't match the signatures

src/diffusers/models/autoencoders/autoencoder_kl_hunyuanimage_refiner.py

@@ -846,13 +846,9 @@ def tiled_decode(self, z: torch.Tensor) -> torch.Tensor:
 
         Args:
             z (`torch.Tensor`): Input batch of latent vectors.
-            return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`~models.vae.DecoderOutput`] instead of a plain tuple.
 
         Returns:
-            [`~models.vae.DecoderOutput`] or `tuple`:
-                If return_dict is True, a [`~models.vae.DecoderOutput`] is returned, otherwise a plain `tuple` is
-                returned.
+            `torch.Tensor`: The decoded images.
         """
 
         _, _, _, height, width = z.shape

src/diffusers/models/autoencoders/autoencoder_kl_hunyuanvideo15.py

@@ -880,13 +880,9 @@ def tiled_decode(self, z: torch.Tensor) -> torch.Tensor:
 
         Args:
             z (`torch.Tensor`): Input batch of latent vectors.
-            return_dict (`bool`, *optional*, defaults to `True`):
-                Whether or not to return a [`~models.vae.DecoderOutput`] instead of a plain tuple.
 
         Returns:
-            [`~models.vae.DecoderOutput`] or `tuple`:
-                If return_dict is True, a [`~models.vae.DecoderOutput`] is returned, otherwise a plain `tuple` is
-                returned.
+            `torch.Tensor`: The decoded images.
         """
 
         _, _, _, height, width = z.shape

src/diffusers/models/embeddings.py

@@ -1091,8 +1091,10 @@ def get_2d_rotary_pos_embed_lumina(embed_dim, len_h, len_w, linear_factor=1.0, n
     Args:
     embed_dim: (`int`):
         The embedding dimension size, corresponding to hidden_size_head.
-    grid (`np.ndarray`):
-        The grid of the positional embedding.
+    len_h (`int`):
+        The height of the grid of the positional embedding.
+    len_w (`int`):
+        The width of the grid of the positional embedding.
     linear_factor (`float`):
         The linear factor of the positional embedding, which is used to scale the positional embedding in the linear
         layer.

src/diffusers/models/transformers/transformer_temporal.py

@@ -288,8 +288,6 @@ def forward(
         Args:
             hidden_states (`torch.Tensor` of shape `(batch size, channel, height, width)`):
                 Input hidden_states.
-            num_frames (`int`):
-                The number of frames to be processed per batch. This is used to reshape the hidden states.
             encoder_hidden_states ( `torch.LongTensor` of shape `(batch size, encoder_hidden_states dim)`, *optional*):
                 Conditional embeddings for cross attention layer. If not given, cross-attention defaults to
                 self-attention.