Abstract
Many organizations fine-tune publicly available pretrained Automatic Speech Recognition (ASR) models and deploy them in black-box settings, assuming limited access provides protection. We show this assumption is fragile: adversarial perturbations crafted on the public base model transfer effectively to fine-tuned target models, severely degrading performance and posing concerns for safety-critical applications. We propose TransferBreaker, a unified fine-tuning framework that suppresses adversarial transfer by integrating Base Adversarial Fine-Tuning, which restricts adversarial training to base-effective perturbations; Latent Jacobian Regularization, which enforces latent-space invariance by suppressing adversarially sensitive directions; and HybridGrad-AFT, which improves robustness against adaptive attacks by interpolating transferable perturbations from base and target gradients. We theoretically justify all components and evaluate TransferBreaker across three languages and four large ASR models, reducing adversarial WER from 92.6 to 27.8. Our code is publicly available at https://github.com/rohban-lab/TransferBreaker.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Nafez, M., Mousavi, A., Mahdavi, M. E., Poulaei, M., Feriz, K. K., & Rohban, M. H. (2026). Breaking Adversarial Transferability in Fine-Tuned Speech Recognition. https://omanscience.com/en/articles/breaking-adversarial-transferability-in-fine-tuned-speech-recognition
MLA 9
Nafez, Mojtaba, et al. "Breaking Adversarial Transferability in Fine-Tuned Speech Recognition." https://omanscience.com/en/articles/breaking-adversarial-transferability-in-fine-tuned-speech-recognition.
Chicago (author–date)
Nafez, Mojtaba, Aref Mousavi, Mohammad Ebrahim Mahdavi, Mobina Poulaei, Kiarash Kiani Feriz, and Mohammad Hossein Rohban. 2026. "Breaking Adversarial Transferability in Fine-Tuned Speech Recognition." https://omanscience.com/en/articles/breaking-adversarial-transferability-in-fine-tuned-speech-recognition.
Harvard
Nafez, M., Mousavi, A., Mahdavi, M. E., Poulaei, M., Feriz, K. K. and Rohban, M. H. (2026) 'Breaking Adversarial Transferability in Fine-Tuned Speech Recognition', Available at: https://omanscience.com/en/articles/breaking-adversarial-transferability-in-fine-tuned-speech-recognition.
Vancouver
Nafez M, Mousavi A, Mahdavi ME, Poulaei M, Feriz KK, Rohban MH. Breaking Adversarial Transferability in Fine-Tuned Speech Recognition. https://omanscience.com/en/articles/breaking-adversarial-transferability-in-fine-tuned-speech-recognition
IEEE
M. Nafez, A. Mousavi, M. E. Mahdavi, M. Poulaei, K. K. Feriz, and M. H. Rohban, "Breaking Adversarial Transferability in Fine-Tuned Speech Recognition," https://omanscience.com/en/articles/breaking-adversarial-transferability-in-fine-tuned-speech-recognition.