Abstract
Audio deepfake detection (ADD) must remain effective when new spoofing attacks emerge after deployment. Emerging audio language model (ALM)-based ADD methods are built on predefined supervision from ground-truth labels or verified forensic rationales. However, this paradigm overlooks an ALM's own mistakes, which indicate where targeted supervision is most needed. To this end, we first introduce evolving spoofing environments for ALM-based ADD, where a new attack becomes dominant while previously observed attacks persist. Motivated by the above learning-from-mistakes perspective, we further propose SE-ADD, a self-evolving framework that iteratively adapts an ALM via low-rank adaptation (LoRA) using mistake-driven supervision built from its verdicts and self-generated forensic cues. All training samples receive direct authenticity supervision, while misclassified ones receive additional cue-augmented supervision. As verdicts and cues are regenerated by the updated ALM, the resulting supervision evolves accordingly. Experiments on two ALMs demonstrate the effectiveness of SE-ADD in generalizing to unseen attacks, reducing the equal error rate (EER) from $36.72\%$ to $7.52\%$ for Qwen2-Audio and from $19.93\%$ to $3.97\%$ for MOSS-Audio.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Wan, R., Xie, W., Li, J., Wang, W., Yin, L., Song, Y., & Wang, X. (2026). SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision. https://omanscience.com/en/articles/se-add-self-evolving-audio-deepfake-detection-with-mistake-driven-supervision
MLA 9
Wan, Rong, et al. "SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision." https://omanscience.com/en/articles/se-add-self-evolving-audio-deepfake-detection-with-mistake-driven-supervision.
Chicago (author–date)
Wan, Rong, Wei Xie, Jiaxi Li, Wenwu Wang, Lu Yin, Yiliao Song, and Xilu Wang. 2026. "SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision." https://omanscience.com/en/articles/se-add-self-evolving-audio-deepfake-detection-with-mistake-driven-supervision.
Harvard
Wan, R., Xie, W., Li, J., Wang, W., Yin, L., Song, Y. and Wang, X. (2026) 'SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision', Available at: https://omanscience.com/en/articles/se-add-self-evolving-audio-deepfake-detection-with-mistake-driven-supervision.
Vancouver
Wan R, Xie W, Li J, Wang W, Yin L, Song Y, et al. SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision. https://omanscience.com/en/articles/se-add-self-evolving-audio-deepfake-detection-with-mistake-driven-supervision
IEEE
R. Wan, W. Xie, J. Li, W. Wang, L. Yin, Y. Song, and X. Wang, "SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision," https://omanscience.com/en/articles/se-add-self-evolving-audio-deepfake-detection-with-mistake-driven-supervision.