Copy nii-yamagishilab/xls-r-1b-anti-deepfake for the Arena (CC BY-NC-SA 4.0)

README.md

@@ -0,0 +1,33 @@
+---
+license: cc-by-nc-sa-4.0
+base_model:
+- nii-yamagishilab/xls-r-1b-anti-deepfake
+tags:
+- anti-spoofing
+- audio-deepfake-detection
+- speech
+pipeline_tag: audio-classification
+---
+
+# XLS-R-1B-AntiDeepfake
+
+Bit-exact copy of [`nii-yamagishilab/xls-r-1b-anti-deepfake`](https://huggingface.co/nii-yamagishilab/xls-r-1b-anti-deepfake) (the AntiDeepfake
+collection, nii-yamagishilab), hosted for the Speech Anti-Spoofing **Arena**.
+Trained with RawBoost data augmentation (default).
+
+- **Paper:** Post-training for Deepfake Speech Detection, [arXiv:2506.21090](https://arxiv.org/abs/2506.21090)
+- **Original code:** https://github.com/nii-yamagishilab/AntiDeepfake
+- **Backbone:** XLS-R 1B; head = AdaptiveAvgPool1d + Linear(D, 2). Score = real logit.
+- **License:** CC BY-NC-SA 4.0 (non-commercial) — inherited from the source.
+
+## ⚠️ Evaluation restriction
+
+It is **STRICTLY PROHIBITED** to evaluate this model on its training / in-domain
+corpora: ASVspoof2019 LA, ASVspoof2021 LA/DF, ASVspoof5, CFAD, DECRO, DFADD,
+"Diffuse or Confuse", DiffSSD, DSD, HABLA, MLAAD, SpoofCeleb, CVoiceFake, any
+LibriTTS-based set (incl. SONAR, CD-ADD, LibriSeVoc), any VoxCeleb-based set,
+WaveFake, any FLEURS-based set, CodecFake (any variant), any AIShell3-based set,
+and any MLS-based set. On the Arena it is evaluated **only** on ArAD and
+In-the-Wild.
+
+All credit to the original authors (Ge, Wang, et al., NII).

config.json

@@ -0,0 +1,109 @@
+{
+  "_name_or_path": "xls-r-1b-anti-deepfake",
+  "activation_dropout": 0.0,
+  "adapter_attn_dim": null,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForPreTraining"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 1024,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "sum",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.1,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 1280,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.075,
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 48,
+  "num_negatives": 100,
+  "output_hidden_size": 1280,
+  "pad_token_id": 0,
+  "proj_codevector_dim": 1024,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.2",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32,
+  "xvector_output_dim": 512
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57da6a47234bb8aad0c61b7f575378a12c2734a30f1f395b95bbf97ed4ba5373
+size 3862169656