Remove deprecated import, type hint

modeling_qwen3_shared_moe.py

@@ -20,17 +20,12 @@ import torch
 from torch import nn
 import torch.nn.functional as F
 
-from transformers.modeling_flash_attention_utils import FlashAttentionKwargs
 from transformers.modeling_outputs import (
     MoeCausalLMOutputWithPast,
     MoeModelOutputWithPast,
 )
 from transformers.activations import ACT2FN
-from transformers.processing_utils import Unpack
-from transformers.utils import (
-    LossKwargs,
-    logging,
-)
+from transformers.utils import logging
 from transformers.models.mixtral.modeling_mixtral import (
     load_balancing_loss_func,
 )
@@ -132,9 +127,6 @@ class Qwen3SharedMoeModel(Qwen3MoeModel):
         )
 
 
-class KwargsForCausalLM(FlashAttentionKwargs, LossKwargs): ...
-
-
 class Qwen3SharedMoeForCausalLM(Qwen3MoeForCausalLM):
     config_class = Qwen3SharedMoeConfig
 
@@ -157,7 +149,7 @@ class Qwen3SharedMoeForCausalLM(Qwen3MoeForCausalLM):
         output_router_logits: Optional[bool] = None,
         cache_position: Optional[torch.LongTensor] = None,
         logits_to_keep: Union[int, torch.Tensor] = 0,
-        **kwargs: Unpack[KwargsForCausalLM],
+        **kwargs,
     ) -> MoeCausalLMOutputWithPast:
         r"""
             labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):