用量子启发结构提升小样本生成内容检测效果
Quantum-Inspired Fine-Tuning for Few-Shot AIGC Detection via Phase-Structured Reparameterization
- 将量子神经网络的相位结构融入LoRA适配器，实现轻量微调
- 在少样本场景下比标准LoRA高5%以上准确率
- 提出纯经典版本H-LoRA，成本更低且性能相近
近期研究表明，量子神经网络（QNN）在小样本场景下泛化能力优异。为将此优势扩展至大规模任务，我们提出Q-LoRA——一种将轻量级QNN集成到低秩适配（LoRA）适配器中的量子增强微调方案。应用于人工智能生成内容（AIGC）检测时，Q-LoRA在少样本设置下持续优于标准LoRA。我们分析其提升来源，发现两个可能的结构归纳偏置：(i) 相位感知表示，可编码正交幅度-相位分量中的更丰富信息；(ii) 范数约束变换，通过固有正交性稳定优化过程。然而，Q-LoRA因量子模拟带来显著开销。基于此分析，我们进一步引入H-LoRA，一种完全经典的变体，在LoRA适配器中引入希尔伯特变换以保留相似相位结构与约束。在少样本AIGC检测实验中，两者均较标准LoRA提升超5%准确率，且H-LoRA在显著更低开销下达到相当性能。
原文摘要 · Abstract (English)
Recent studies show that quantum neural networks (QNNs) generalize well in few-shot regimes. To extend this advantage to large-scale tasks, we propose Q-LoRA, a quantum-enhanced fine-tuning scheme that integrates lightweight QNNs into the low-rank adaptation (LoRA) adapter. Applied to AI-generated content (AIGC) detection, Q-LoRA consistently outperforms standard LoRA under few-shot settings. We analyze the source of this improvement and identify two possible structural inductive biases from QNNs: (i) phase-aware representations, which encode richer information across orthogonal amplitude-phase components, and (ii) norm-constrained transformations, which stabilize optimization via inherent orthogonality. However, Q-LoRA incurs non-trivial overhead due to quantum simulation. Motivated by our analysis, we further introduce H-LoRA, a fully classical variant that applies the Hilbert transform within the LoRA adapter to retain similar phase structure and constraints. Experiments on few-shot AIGC detection show that both Q-LoRA and H-LoRA outperform standard LoRA by over 5% accuracy, with H-LoRA achieving comparable accuracy at significantly lower cost in this task.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。