这项研究揭示了训练策略对模型鲁棒性的意外影响，做语音反欺骗的团队可以直接用RAT提升检测性能，值得关注。
该研究提出了一种基于说话人参考录音的条件反欺骗架构，但发现模型在推理时会忽略参考信息。令人意外的是，使用参考通道进行训练能诱导模型产生不变性，从而提升深度伪造检测能力，即使推理时参考缺失或不匹配。基于此，研究者提出了参考增强训练（RAT）策略，在推理时即使将参考替换为零向量，检测性能仍优于单次语音基线。通过严格分析，他们证明优化过程会迅速降低参考贡献，使推理几乎独立于参考通道。采用RAT，单个检测器在ASVspoof 5基准上实现了2.57%的等错误率和0.074的最小检测代价函数，超越了大型集成系统。
RAT: Reference-Augmented Training for ASV Anti-Spoofing
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. Surprisingly, training with a reference channel induces invariance that improves deepfake detection, even when the reference is absent or mismatched during inference. Based on this observation, we propose a Reference-Augmented Training (RAT) strategy. RAT yields improved detection performance compared to single-utterance baselines, even when the reference recording is replaced with a zero vector at inference. Through rigorous analysis, we demonstrate that the optimization process rapidly diminishes the reference contributions, leading to inference largely independent of the reference channel. Using RAT, we achieve state-of-the-art 2.57% EER and 0.074 minDCF on the ASVspoof 5 benchmark with a single detector, surpassing even large ensemble systems.