AIBullisharXiv – CS AI · Mar 45/103
🧠
Quantum-Inspired Fine-Tuning for Few-Shot AIGC Detection via Phase-Structured Reparameterization
Researchers propose Q-LoRA, a quantum-enhanced fine-tuning method that integrates quantum neural networks into LoRA adapters for improved AI-generated content detection. The study also introduces H-LoRA, a classical variant using Hilbert transforms that achieves similar 5%+ accuracy improvements over standard LoRA at lower computational cost.