发表机构
University of Stuttgart; National Institute of Informatics; German Research Center for Artificial Intelligence (DFKI); Technical University of Berlin（斯图加特大学; 日本信息处理研究所; 德国人工智能研究中心; 柏林技术大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究如何将自我监督学习模型用于音频深度伪造检测，提出证据子空间投影方法，通过在共享空间表示证据因素和真实性标签，投影决策向量量化证据解释力，经多数据集多设置评估，验证方法并获新见解。
AI 中文摘要
自我监督学习（SSL）模型被广泛用作最先进的音频深度伪造检测的特征提取器，但尚不清楚如何直接和定量地将SSL模型捕获的内容与检测决策联系起来。为填补这一空白，我们提出证据子空间投影方法，该方法在由SSL模型的神经元激活模式构建的共享空间中表示证据因素（如攻击类别、编解码器、性别、传输）和真实性标签。通过将决策向量投影到每个证据子空间上，我们获得一个标量比率，量化每种证据类型的解释力。我们在多个数据集上对原始、微调及后训练设置下的SSL模型进行评估。结果证实了现有研究的发现，验证了所提方法，并揭示了模型行为的新见解。
英文摘要
Self-supervised learning (SSL) models are widely used as feature extractors for state-of-the-art audio deepfake detection, but it remains unclear how to directly and quantitatively connect what SSL models capture to detection decisions. To address this gap, we propose Evidence Subspace Projection, a method that represents both evidence factors (e.g., attack category, codec, gender, transmission) and authenticity labels in a shared space constructed from SSL models' neuron activation patterns. By projecting the decision vector onto each evidence subspace, we obtain a scalar ratio that quantifies the explanatory power of each evidence type. We evaluate SSL models in raw, fine-tuned, and post-trained settings on multiple datasets. The results confirm findings from established studies, validating the proposed method, and reveal new insights into model behavior.
CommentsAccepted to Interspeech 2026
