1
The Perceived Fragility of Explanations in Audio Models: Manipulation of Attribution with Unchanged Predictions
音频模型解释并非可信：研究发现可在不改变预测结果的情况下操纵归因，揭示AI可解释性的潜在脆弱性。
arXiv:2606.14466v1 Announce Type: cross Abstract: This paper investigates the fragility of post-hoc explanation methods in audio deepfake detection. W…