发表机构
Victoria University; CSIRO; Australian International Institute of Higher Education（维多利亚大学; 联邦科学与工业研究组织; 澳大利亚国际高等教育学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过10量子比特混合变分电路在三个基准上验证参数高效量子NLP，实现超越经典基线并展现对抗鲁棒性，揭示多量子比特纠缠为性能关键。
AI 中文摘要
量子机器学习在自然语言任务上的严格实证验证仍然稀缺。我们评估了一个10量子比特的混合量子-经典变分电路（2,148个参数）在三个基准上的释义检测性能：MRPC、Quora问题对（QQP）和对抗性PAWS。在QQP上（n=10个种子），该电路达到75.53%±0.75%的准确率，统计上优于参数匹配的经典基线（DeepMLP：p=0.015，Cohen's d=1.20；F1：p<0.001，d=2.32），并以少31,191个参数超越DistilBERT-4bit。在MRPC上，最优的2层变体以低54,000个参数的成本达到BERT-base准确率的92%。电路深度分析揭示了数据集-深度缩放效应；通过Meyer-Wallach度量的纠缠分析识别出多量子比特纠缠是主要性能驱动因素（四个变体间r=0.85）。在PAWS上的对抗性评估揭示了涌现鲁棒性：98.2%的召回率对比经典的81.6%（+16.6个百分点，d=1.24，p<0.001），且无需对抗训练。这些结果构成了混合变分电路在NLP中首次系统的参数匹配多基准实证验证。所有结果均来自经典模拟；硬件验证是未来工作。
英文摘要
Rigorous empirical validation of quantum machine learning on natural language tasks remains scarce. We evaluate a 10-qubit hybrid quantum-classical variational circuit (2,148 parameters) for paraphrase detection across three benchmarks: MRPC, Quora Question Pairs (QQP), and adversarial PAWS. On QQP (n = 10 seeds), the circuit achieves 75.53% +/- 0.75%. accuracy, statistically outperforming parameter-matched classical baselines (DeepMLP: p = 0.015, Cohen's d = 1.20; F1: p < 0.001, d = 2.32) and surpassing DistilBERT-4bit with 31,191 fewer parameters. On MRPC the optimal 2-layer variant reaches 92% of BERT-base accuracy at 54,000 lower parameter cost. Circuit depth analysis reveals a dataset-depth scaling effect; entanglement analysis via the Meyer-Wallach measure identifies multi-qubit entanglement as the primary performance driver (r = 0.85 across four variants). Adversarial evaluation on PAWS reveals emergent robustness: 98.2% recall versus 81.6% classical (+16.6 pp, d = 1.24, p < 0.001), without adversarial training. These results constitute the first systematic parameter-matched multi-benchmark empirical validation of hybrid variational circuits for NLP. All results are from classical simulation; hardware validation is future work.
