Add DeepFense-CodecFake-W2V2-MLP submission for DECRO (EER=11.1119)

submissions/deepfense-codecfake-w2v2-mlp.yaml

@@ -0,0 +1,59 @@
+schema_version: 4
+system:
+  name: DeepFense-CodecFake-W2V2-MLP
+  slug: deepfense-codecfake-w2v2-mlp
+  description: 'DeepFense (Kheir et al., 2026) CodecFake recipe: a fully fine-tuned wav2vec 2.0 XLS-R
+    300M front-end (last transformer layer, features_only) followed by an MLP back-end (Linear 1024->128,
+    LayerNorm, ReLU, mean pooling over time) and a 2-way linear CrossEntropy head. Trained on the Codecfake
+    corpus (Xie et al., AISHELL-3/VCTK-based codec re-synthesis) without data augmentation, seed 42; the
+    released best_model.pth (epoch 2) as is. Input as the DeepFense eval pipeline: raw float waveform
+    (no normalisation), first 64000 samples (4 s), shorter clips repeat-padded. score = logit(bonafide)
+    - logit(spoof); FP32.'
+  code: https://github.com/Yaselley/deepfense-framework/tree/3f2fe1146c48ea4bacc480dd96d806b6c72c09cf
+  checkpoint: https://huggingface.co/DeepFense/CodecFake_Wav2Vec2_MLP_NoAug_Seed42/blob/c08bd2125c83e6c8fdc64566bfbfda18dff4dfb4/best_model.pth
+  params_millions: 317.522306
+  paper:
+    arxiv_id: '2604.08450'
+    url: https://arxiv.org/abs/2604.08450
+    bibtex: "@article{kheir2026deepfense,\n  title={DeepFense: A Unified, Modular, and Extensible Framework\
+      \ for Robust Deepfake Audio Detection},\n  author={Kheir, Yassine El and Das, Arnab and Xiao, Yixuan\
+      \ and Wang, Xin and Kallel, Feidi and Erdogan, Enes Erdem and Vu, Ngoc Thang and Polzehl, Tim and\
+      \ Moeller, Sebastian},\n  journal={arXiv preprint arXiv:2604.08450},\n  year={2026}\n}\n"
+dataset:
+  id: SpeechAntiSpoofingBenchmarks/DECRO
+  revision: e54c29df5b81bac75e5cc294d1ab0d4db72a5c71
+  split: test
+scores:
+  eer_percent: 11.111937246737796
+  n_trials: 37314
+  n_skipped: 0
+artifact:
+  scores_url: https://huggingface.co/SpeechAntiSpoofingBenchmarks/DeepFense-CodecFake-W2V2-MLP/resolve/2743829e6404abf5006cc8bf279edb56901a1434/.eval_results/SpeechAntiSpoofingBenchmarks/DECRO/scores.txt
+  scores_sha256: 4a0fc3b5f67de72022f088114ba0450943f301b78718ab90de74268aeb977a4c
+  bench_version: speech-spoof-bench==0.4.1
+reproduction:
+  reproduced_by: SpeechAntiSpoofingBenchmarks
+  reproduced_at: '2026-11-07'
+  reproduced_bench_version: speech-spoof-bench==0.4.1
+  match: inference
+submitter:
+  hf_username: korallll
+  contact: k.n.borodin@mtuci.ru
+submitted_at: '2026-11-07'
+notes: 'DeepFense-CodecFake-W2V2-MLP = HF DeepFense/CodecFake_Wav2Vec2_MLP_NoAug_Seed42 @ c08bd2125c83e6c8fdc64566bfbfda18dff4dfb4,
+  best_model.pth (sha256 1d4bf44afd0f7e887266c8dc49875b0ddb60a07fb16279fa2d56598519ea3c1a; a training
+  checkpoint with model_state + Adam optimizer_state, epoch 2, step 46298). The HF repo states no weight
+  license, so the weights are linked, not re-hosted; the wrapper loads only model_state, saved unchanged
+  as model_state.safetensors (sha256 5c97bd4ae3a2075aa8df309bde8b7dd7f9ce540a50df969d4ff0ba7a74a560a5,
+  435 fp32 tensors). Code: Yaselley/deepfense-framework @ 3f2fe11 (Apache-2.0) + the checkpoint''s config.yaml
+  (StandardDetector: wav2vec2 fairseq frontend on xlsr2_300m.pt, MLP backend projection [128], layer norm,
+  ReLU, mean pooling; CrossEntropy n_classes 2; label_map bonafide=1, spoof=0). XLS-R 300M is the fairseq
+  Wav2Vec2Model built from the xlsr2_300m.pt model config; all 435 checkpoint tensors load strictly (the
+  7 pre-training quantizer/project_q/final_proj tensors are unused on the features_only path). Eval preprocessing
+  = StandardDataset + pad: soundfile float mono, first 64000 samples, np.tile repeat-pad, no normalisation;
+  no resampling; FP32. Score = logit1 - logit0 (higher = bona fide). The authors'' test.py EER uses get_score
+  = logit1 only (a different, non-monotone-equivalent score); see implementation-notes.md. Verified against
+  the authors'' ModularDetector + fairseq_local loader on 12 clips: max |d logits| 4.8e-6. Training: Codecfake,
+  no augmentation, seed 42. Self-reported EER (DeepFense deepfense-results Space): Codecfake 10.71, In-the-Wild
+  8.10, ASVspoof2019 LA 4.15, ASVspoof2021 DF 6.10, ASVspoof2021 LA 15.46, ASVspoof5 29.93, ODSS 11.51,
+  HABLA 7.82, ADD22 track 3 7.48.'