Multi-Backbone Self-Supervised Ensembles for Audio Deepfake Detection and a Cross-Track Analysis of Generation-Detection Asymmetry

• URL: http://arxiv.org/abs/2608.01796v1
• Date: Mon, 03 Aug 2026 07:04:51 GMT
• Title: Multi-Backbone Self-Supervised Ensembles for Audio Deepfake Detection and a Cross-Track Analysis of Generation-Detection Asymmetry
• Abstract summary: This paper describes the participation of team "Go-To-Germany" in the ImageCLEF 2026 Audio Deepfake Detection and Generation task.<n>We present a cross-track analysis revealing a pronounced asymmetry: our detection system identifies 100% of participant-generated deepfakes.<n>We report an 11.25% false-positive gap on held-out organizer real recordings as the principal open challenge for deployment.
• Score: 9.42962723864652
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper describes the participation of team "Go-To-Germany" in the ImageCLEF 2026 Audio Deepfake Detection and Generation task. Our detection system, built on a four-backbone self-supervised learning (SSL) ensemble combining WavLM-Large, Wav2Vec2-XLS-R-300M, ECAPA-TDNN, and x-vector representations, achieved a final score of 0.9522 on the official ImageCLEF 2026 evaluation, with perfect accuracy (1.0000) on participant-generated deepfakes and 0.8875 on the held-out organizer ground-truth real data. For the Generation sub-task, our official team submission, an F5-TTS v1 baseline processed with a uniform reverberation pass and submitted as a deliberate anti-forensic probe, ranked first with a final score of 0.4304 (word error rate (WER) 4.99%, character error rate (CER) 2.07%); details of our four-model program (GLM-TTS, F5-TTS, XTTS v2, CosyVoice3), from which the official entry was drawn, appear in the paper. We present a cross-track analysis revealing a pronounced asymmetry: our detection system identifies 100% of participant-generated deepfakes, while our official generation entry, despite ranking first in the Audio Generation sub-task and evading 61.4% and 56.2% of participant and organizer detectors, attains a Final Score of 0.4304 against 0.9522 on the Detection side. We further report falsification-based ablation experiments (LOSO 56-speaker cross-validation, three-region backbone geometry, bootstrap confidence intervals, and PCA analysis) that motivate our architectural-insurance hypothesis for multi-backbone SSL ensembling. We complement these results with five cross-track insights and five pre-registered falsification experiments connecting generation-side evasion to detection-side design decisions, and we openly report an 11.25% false-positive gap on held-out organizer real recordings as the principal open challenge for deployment.

Related papers

• Comparing Self-Supervised and Domain-Invariant Features for Cross-Domain Voice Phishing Detection [5.314238974401925]
  Domain-invariant prosodic features and self-supervised representations are compared.<n>HuBERT achieves highest performance (94.2% F1, 5-shot), while wav2vec2.0 exhibits a precision-oriented detection profile (90.2% F1 with 99.4% precision, 5-shot)<n>These findings reveal fundamental trade-offs: domain-invariant features enable zero-shot deployment when no real data exists, while SSL methods achieve higher performance but require real samples and compute.
  arXiv  Detail & Related papers  (2026-09-07T06:06:45Z)
• Adversarial Deepfake Generation and an Investigation of Purification-Based Adversarial Detection [9.42962723864652]
  This paper describes the participation of team "Go To Germany" in the ImageCLEF 2026 Deepfake Detection and Generation Task.<n>For the image generation task, we employ FLUX.1-dev with PuLID for identity-preserving face synthesis.<n>For the image detection task, we combine two complementary detectors - SigLIP+DINOv2 for AI-generated images and GenD-DINOv3 for face manipulations.
  arXiv  Detail & Related papers  (2026-07-28T15:19:16Z)
• SingGuard-NSFA: Extensible Guardrails for Agentic AI via Generative Reasoning and Real-Time Classification [0.0]
  nsfaguard is a guardrail framework for securing agentic AI systems against operational threats.<n>We first introduce the NSFA taxonomy, which organizes 185 risk variants into a CIA-triad-grounded hierarchy.<n>We develop a dual-mode approach combining SFT-based generative reasoning for interpretable offline auditing with discriminative classification heads on the frozen backbone.
  arXiv  Detail & Related papers  (2026-07-13T07:28:45Z)
• Efficient Visual Pointing for Embodied AI:Agent-Driven Data Synthesis, Cross-Block Attention, and Iterative Correction [55.11480729304395]
  We describe our PointArena 2026 solution, which achieves 77.2% overall accuracy and ranks second on the benchmark.<n>The ap proach targets three failure modes. First, agent-driven syn thesis builds large semantic and anchor-relative candidate pools.<n>Second, a determinis tic steerable-data pipeline creates a verified 10,000-sample main set, plus reserve samples, using masks, templates, and path verification.
  arXiv  Detail & Related papers  (2026-06-29T06:39:03Z)
• Explainable Deep Learning for Brain Tumor Classification: Comprehensive Benchmarking with Dual Interpretability and Lightweight Deployment [4.259927630334864]
  This study provides a full deep learning system for automated classification of brain tumors from MRI images.<n>Inception-ResNet V2 reached state-of-the-art performance, achieving a 99.53% accuracy on testing.<n>This end-to-end solution considers accuracy, interpretability, and deployability of trustworthy AI.
  arXiv  Detail & Related papers  (2025-11-20T17:21:40Z)
• MIRAGE: Agentic Framework for Multimodal Misinformation Detection with Web-Grounded Reasoning [0.6475163438744868]
  We present MIRAGE, an inference-time, model-pluggable agentic framework that decomposes multimodal verification into four sequential modules.<n> visual veracity assessment detects AI-generated images, cross-modal consistency analysis identifies out-of-context repurposing, retrieval-augmented factual checking grounds claims in web evidence.<n>MIRAGE orchestrates vision-language model reasoning with targeted web retrieval, outputs structured and citation-linked rationales.
  arXiv  Detail & Related papers  (2025-10-20T14:40:26Z)
• When Fine-Tuning is Not Enough: Lessons from HSAD on Hybrid and Adversarial Audio Spoof Detection [3.7411108810335922]
  Spoof detection is a challenge for voice authentication, smart assistants, and telecom security.<n>We present a benchmark containing 1,248 clean and 41,044 degraded utterances across four classes: human, cloned, zero-shot AI-generated, and hybrid audio.<n>Results reveal critical lessons: pretrained models overgeneralize and collapse under hybrid conditions; spoof-specific fine-tuning improves separability but struggles with unseen compositions.
  arXiv  Detail & Related papers  (2025-09-09T01:43:28Z)
• Lie Detector: Unified Backdoor Detection via Cross-Examination Framework [68.45399098884364]
  We propose a unified backdoor detection framework in the semi-honest setting.<n>Our method achieves superior detection performance, improving accuracy by 5.4%, 1.6%, and 11.9% over SoTA baselines.<n> Notably, it is the first to effectively detect backdoors in multimodal large language models.
  arXiv  Detail & Related papers  (2025-03-21T06:12:06Z)
• Whole-body Detection, Recognition and Identification at Altitude and Range [57.445372305202405]
  We propose an end-to-end system evaluated on diverse datasets. Our approach involves pre-training the detector on common image datasets and fine-tuning it on BRIAR's complex videos and images. We conduct thorough evaluations under various conditions, such as different ranges and angles in indoor, outdoor, and aerial scenarios.
  arXiv  Detail & Related papers  (2023-11-09T20:20:23Z)
• Scenario Aware Speech Recognition: Advancements for Apollo Fearless Steps & CHiME-4 Corpora [70.46867541361982]
  We consider a general non-semantic speech representation, which is trained with a self-supervised criteria based on triplet loss called TRILL. We observe +5.42% and +3.18% relative WER improvement for the development and evaluation sets of Fearless Steps.
  arXiv  Detail & Related papers  (2021-09-23T00:43:32Z)
• G-DetKD: Towards General Distillation Framework for Object Detectors via Contrastive and Semantic-guided Feature Imitation [49.421099172544196]
  We propose a novel semantic-guided feature imitation technique, which automatically performs soft matching between feature pairs across all pyramid levels. We also introduce contrastive distillation to effectively capture the information encoded in the relationship between different feature regions. Our method consistently outperforms the existing detection KD techniques, and works when (1) components in the framework are used separately and in conjunction.
  arXiv  Detail & Related papers  (2021-08-17T07:44:27Z)
• FoolHD: Fooling speaker identification by Highly imperceptible adversarial Disturbances [63.80959552818541]
  We propose a white-box steganography-inspired adversarial attack that generates imperceptible perturbations against a speaker identification model. Our approach, FoolHD, uses a Gated Convolutional Autoencoder that operates in the DCT domain and is trained with a multi-objective loss function. We validate FoolHD with a 250-speaker identification x-vector network, trained using VoxCeleb, in terms of accuracy, success rate, and imperceptibility.
  arXiv  Detail & Related papers  (2020-11-17T07:38:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.