← Back to all news
arXiv AI

CoReLoop: Parameter-Efficient Controlled Recurrent Refinement for Audio Deepfake Detection

CoReLoop introduces a parameter‑efficient refinement strategy for audio deepfake detection that reuses a frozen SSL‑based detector’s encoder outputs without altering its original parameters. By adapting recurrent inputs, controlling state updates, and aligning refined outputs with the frozen classifier, the method adds lightweight refinement modules and low‑rank adapters trained on the original data. On 14 cross‑domain test sets, the 24‑layer model reduces pooled equal error rate from 4.85% to 3.74% with two passes, and an optional halting head further improves performance to 3.73% with an average of 1.18 passes.

Related stories

arXiv AI
Sep 18

CoRELoop: Parameter-Efficient Controlled Recurrent Refinement for Audio Deepfake Detection

CoRELoop introduces a parameter‑efficient refinement framework for audio deepfake detection that operates on a pre‑trained SSL‑based detector without altering its original parameters. By adapting recurrent inputs, controlling state updates, and aligning refined outputs with the frozen classifier, CoRELoop adds lightweight refinement modules and low‑rank adapters, achieving a pooled equal error rate reduction from 4.85% to 3.74% on 14 cross‑domain test sets with only about 10 M trainable parameters. An optional halting head further optimizes performance, reaching 3.73% pooled EER with an average of 1.18 passes.

By Kunyu Feng, Yuxiang Wang, Li Wang, Wan Lin, Zhizheng Wu
More like this →
arXiv AI
Aug 17

Teffic-Audio: Tell Fact from Fiction

arXiv:2607. 28351v2 Announce Type: replace-cross Abstract: Speech deepfake detection has expanded in scope with increasingly heterogeneous spoofing mechanisms, including speech synthesis, voice conversion, vocoder reconstruction, and neural-codec resynthesis.

By Wan Lin, Li Wang, Jindong Wang, Kunyu Feng, Zhizheng Wu
benchmarks
More like this →
arXiv AI
Jun 10

RAT: Reference-Augmented Training for ASV Anti-Spoofing

arXiv:2606. 10908v1 Announce Type: cross Abstract: We introduce a spoofing countermeasure architecture conditioned on speaker-reference recordings, but observe that it converges to a solution that effectively ignores the reference during inference.

By Vojt\v{e}ch Stan\v{e}k, Anton Firc, Jakub Re\v{s}, Kamil Malinka
benchmarks
More like this →
arXiv Machine Learning
Sep 15

Parameter isolation with domain-specific experts for incremental audio classification

The paper introduces a domain‑specific parameter‑isolation architecture for domain‑incremental learning (DIL) in audio classification, aiming to preserve knowledge from earlier domains without accessing their data. By employing data‑free generative replay and cross‑domain feature generation, the method constructs new experts conditioned on all previously frozen models, thereby mitigating catastrophic forgetting. Applied to the DCASE 2026 Challenge Task 7, the approach achieves micro and macro accuracies of 78.4 % and 78.9 %, outperforming the baseline by 33 and 25 percentage points, respectively, with ablation studies confirming the contribution of each component.

By Jongyeon Park, Do-Hyeon Lim, Sang-won Park, Hong Kook Kim, Kyungdeuk Ko, Hyeongcheol Geum, Jeong Eun Lim
benchmarks
More like this →