图注意力设计选择至关重要：LoRA自适应音频防欺骗的受控研究
Graph Attention Design Choices Matter: A Controlled Study of LoRA-Adapted Audio Anti-Spoofing
浏览论文内容
中文总结 AI 辅助
本研究通过受控实验分解图注意力层为三个设计维度，发现可学习温度分支显著提升音频防欺骗性能，且设计维度间存在非加性交互。
中文摘要 AI 辅助
音频防欺骗系统日益将自监督学习、参数高效微调和基于图注意力的后端相结合。然而，此类系统的性能提升往往与骨干网络、微调策略和训练协议的同时变化纠缠在一起，使得图注意力设计的独立贡献难以分离。为解决此问题，我们在统一实验设置下对图注意力层进行了系统的受控研究。我们将该层分解为三个可独立测试的设计维度：评分对称性、温度可学习性和路由粒度。这三个维度分别实例化为基于拼接的评分分支、具有可学习温度的LearnT分支和多温度路由分支。每个维度均作为独立门控残差分支实现，从而能够在相同实验设置下评估各个变体及其组合。在五个评估集和五个随机种子上的实验表明，LearnT分支取得了最佳平均等错误率（EER），相对于基线实现了16.1%的相对改进。相比之下，多温度路由分支单独使用并未提升平均性能，但与基于拼接的评分分支结合时，显著降低了跨种子的标准差。此外，两个单独有效的分支同时使用时性能下降，与基线相比相对恶化25.6%。这一发现揭示了图注意力设计维度之间存在强烈的非加性交互。总体而言，结果表明，在参数受限的微调下，图注意力层的改进更多地依赖于容量分配和分支交互，而非简单地增加可学习参数。
英文摘要
Audio anti-spoofing systems increasingly combine self-supervised learning, parameter-efficient fine-tuning, and graph-attention-based backends. However, performance gains in such systems are often entangled with concurrent changes in the backbone, fine-tuning strategy, and training protocol, making the independent contribution of graph attention design difficult to isolate. To address this issue, we conduct a systematic controlled study of the graph attention layer under a unified experimental setting. We decompose the layer into three independently testable design dimensions: scoring symmetry, temperature learnability, and routing granularity. These are instantiated as a concat-based scoring branch, a LearnT branch with learnable temperature, and a multi-temperature routing branch, respectively. Each dimension is implemented as an independently gated residual branch, enabling the evaluation of both individual variants and their combinations under the same experimental setting. Experiments on five evaluation sets with five random seeds show that the LearnT branch achieves the best average equal error rate (EER), yielding a 16.1% relative improvement over the baseline. In contrast, the multi-temperature routing branch does not improve average performance on its own, but substantially reduces cross-seed standard deviation when combined with the concat-based scoring branch. Moreover, two individually effective branches degrade performance when used together, resulting in a 25.6% relative deterioration compared with the baseline. This finding reveals strong non-additive interactions among graph attention design dimensions. Overall, the results suggest that, under parameter-constrained fine-tuning, improvements in graph attention layers depend more on capacity allocation and branch interaction than on simply adding more learnable parameters.
发表机构
- Southwestern University of Finance and Economics（西南财经大学）
- Xiaomi Inc.（小米公司）
- Wuhan University（武汉大学）
- Hubei Luojia Laboratory（湖北珞珈实验室）
机构由 AI 辅助整理，请以论文原文为准。
