Remove redundant argument apply_t5_attn_mask

anima_train.py

@@ -165,7 +165,6 @@ def train(args):
                     args.text_encoder_batch_size,
                     False,
                     False,
-                    args.apply_t5_attn_mask,
                 )
             )
         train_dataset_group.set_current_strategies()
@@ -223,7 +222,6 @@ def train(args):
     # Set text encoding strategy
     caption_dropout_rate = getattr(args, 'caption_dropout_rate', 0.0)
     text_encoding_strategy = strategy_anima.AnimaTextEncodingStrategy(
-        apply_t5_attn_mask=args.apply_t5_attn_mask,
         dropout_rate=caption_dropout_rate,
     )
     strategy_base.TextEncodingStrategy.set_strategy(text_encoding_strategy)
@@ -243,7 +241,6 @@ def train(args):
             args.text_encoder_batch_size,
             args.skip_cache_check,
             is_partial=False,
-            apply_t5_attn_mask=args.apply_t5_attn_mask,
         )
         strategy_base.TextEncoderOutputsCachingStrategy.set_strategy(text_encoder_caching_strategy)
 

anima_train_network.py

@@ -166,7 +166,6 @@ class AnimaNetworkTrainer(train_network.NetworkTrainer):
     def get_text_encoding_strategy(self, args):
         caption_dropout_rate = getattr(args, 'caption_dropout_rate', 0.0)
         self.text_encoding_strategy = strategy_anima.AnimaTextEncodingStrategy(
-            apply_t5_attn_mask=args.apply_t5_attn_mask,
             dropout_rate=caption_dropout_rate,
         )
         return self.text_encoding_strategy
@@ -193,7 +192,6 @@ class AnimaNetworkTrainer(train_network.NetworkTrainer):
                 args.text_encoder_batch_size,
                 args.skip_cache_check,
                 is_partial=False,
-                apply_t5_attn_mask=args.apply_t5_attn_mask,
             )
         return None
 
@@ -471,7 +469,6 @@ class AnimaNetworkTrainer(train_network.NetworkTrainer):
         return train_util.get_sai_model_spec(None, args, False, True, False, is_stable_diffusion_ckpt=True)
 
     def update_metadata(self, metadata, args):
-        metadata["ss_apply_t5_attn_mask"] = args.apply_t5_attn_mask
         metadata["ss_weighting_scheme"] = args.weighting_scheme
         metadata["ss_discrete_flow_shift"] = args.discrete_flow_shift
         metadata["ss_timestep_sample_method"] = getattr(args, 'timestep_sample_method', 'logit_normal')

docs/anima_train_network.md

@@ -169,8 +169,6 @@ Besides the arguments explained in the [train_network.py guide](train_network.md
   - Maximum token length for the Qwen3 tokenizer. Default `512`.
 * `--t5_max_token_length=<integer>`
   - Maximum token length for the T5 tokenizer. Default `512`.
-* `--apply_t5_attn_mask`
-  - Apply attention mask to T5 tokens in the LLM adapter.
 * `--flash_attn`
   - Use Flash Attention for DiT self/cross-attention. Requires `pip install flash-attn`. Falls back to PyTorch SDPA if the package is not installed. Note: Flash Attention is only applied to DiT blocks; the LLM Adapter uses standard attention because it requires attention masks.
 * `--transformer_dtype=<choice>`
@@ -229,7 +227,6 @@ Anima supports 6 independent learning rate groups. Set to `0` to freeze a compon
 * `--sigmoid_scale` - logit_normalタイムステップサンプリングのスケール係数。デフォルト`1.0`。
 * `--qwen3_max_token_length` - Qwen3トークナイザーの最大トークン長。デフォルト`512`。
 * `--t5_max_token_length` - T5トークナイザーの最大トークン長。デフォルト`512`。
-* `--apply_t5_attn_mask` - LLM AdapterでT5トークンにアテンションマスクを適用。
 * `--flash_attn` - DiTのself/cross-attentionにFlash Attentionを使用。`pip install flash-attn`が必要。
 * `--transformer_dtype` - Transformerブロック用の個別dtype。
 
@@ -537,7 +534,6 @@ Anima LoRA学習では、Qwen3テキストエンコーダーのLoRAもトレー
 
 The following Anima-specific metadata is saved in the LoRA model file:
 
-* `ss_apply_t5_attn_mask`
 * `ss_weighting_scheme`
 * `ss_discrete_flow_shift`
 * `ss_timestep_sample_method`
@@ -552,7 +548,6 @@ The following Anima-specific metadata is saved in the LoRA model file:
 
 以下のAnima固有のメタデータがLoRAモデルファイルに保存されます：
 
-* `ss_apply_t5_attn_mask`
 * `ss_weighting_scheme`
 * `ss_discrete_flow_shift`
 * `ss_timestep_sample_method`

library/anima_train_utils.py

@@ -106,11 +106,6 @@ def add_anima_training_arguments(parser: argparse.ArgumentParser):
         default=512,
         help="Maximum token length for T5 tokenizer (default: 512)",
     )
-    parser.add_argument(
-        "--apply_t5_attn_mask",
-        action="store_true",
-        help="Apply attention mask to T5 tokens in LLM adapter",
-    )
     parser.add_argument(
         "--discrete_flow_shift",
         type=float,

library/strategy_anima.py

@@ -86,10 +86,8 @@ class AnimaTextEncodingStrategy(TextEncodingStrategy):
 
     def __init__(
         self,
-        apply_t5_attn_mask: bool = False,
         dropout_rate: float = 0.0,
     ) -> None:
-        self.apply_t5_attn_mask = apply_t5_attn_mask
         self.dropout_rate = dropout_rate
 
     def encode_tokens(
@@ -97,7 +95,6 @@ class AnimaTextEncodingStrategy(TextEncodingStrategy):
         tokenize_strategy: TokenizeStrategy,
         models: List[Any],
         tokens: List[torch.Tensor],
-        apply_t5_attn_mask: Optional[bool] = None,
         enable_dropout: bool = True,
     ) -> List[torch.Tensor]:
         """Encode Qwen3 tokens and return embeddings + T5 token IDs.
@@ -109,8 +106,6 @@ class AnimaTextEncodingStrategy(TextEncodingStrategy):
         Returns:
             [prompt_embeds, attn_mask, t5_input_ids, t5_attn_mask]
         """
-        if apply_t5_attn_mask is None:
-            apply_t5_attn_mask = self.apply_t5_attn_mask
 
         qwen3_text_encoder = models[0]
         qwen3_input_ids, qwen3_attn_mask, t5_input_ids, t5_attn_mask = tokens
@@ -222,10 +217,8 @@ class AnimaTextEncoderOutputsCachingStrategy(TextEncoderOutputsCachingStrategy):
         batch_size: int,
         skip_disk_cache_validity_check: bool,
         is_partial: bool = False,
-        apply_t5_attn_mask: bool = False,
     ) -> None:
         super().__init__(cache_to_disk, batch_size, skip_disk_cache_validity_check, is_partial)
-        self.apply_t5_attn_mask = apply_t5_attn_mask
 
     def get_outputs_npz_path(self, image_abs_path: str) -> str:
         return os.path.splitext(image_abs_path)[0] + self.ANIMA_TEXT_ENCODER_OUTPUTS_NPZ_SUFFIX
@@ -279,7 +272,6 @@ class AnimaTextEncoderOutputsCachingStrategy(TextEncoderOutputsCachingStrategy):
                 tokenize_strategy,
                 models,
                 tokens_and_masks,
-                apply_t5_attn_mask=self.apply_t5_attn_mask,
                 enable_dropout=False,
             )
 

tests/test_anima_cache.py

@@ -200,7 +200,6 @@ def test_text_encoder_cache(args, pairs):
         t5_max_length=args.t5_max_length,
     )
     text_encoding_strategy = AnimaTextEncodingStrategy(
-        apply_t5_attn_mask=False,
         dropout_rate=0.0,
     )
 
@@ -355,7 +354,6 @@ def test_text_encoder_cache(args, pairs):
     # Test drop_cached_text_encoder_outputs
     print(f"\n[2.8] Testing drop_cached_text_encoder_outputs (caption dropout)...")
     dropout_strategy = AnimaTextEncodingStrategy(
-        apply_t5_attn_mask=False,
         dropout_rate=0.5,  # high rate to ensure some drops
     )
     dropped = dropout_strategy.drop_cached_text_encoder_outputs(*stacked)
@@ -401,7 +399,7 @@ def test_full_batch_simulation(args, pairs):
         qwen3_tokenizer=qwen3_tokenizer, t5_tokenizer=t5_tokenizer,
         qwen3_max_length=args.qwen3_max_length, t5_max_length=args.t5_max_length,
     )
-    text_encoding_strategy = AnimaTextEncodingStrategy(apply_t5_attn_mask=False, dropout_rate=0.0)
+    text_encoding_strategy = AnimaTextEncodingStrategy(dropout_rate=0.0)
 
     captions = [cap for _, cap in pairs]