{"id":"arxiv/2609.14529","url":"https://arxiv.org/abs/2609.14529","tags":["quant-ph","cs.CL"],"score":0,"title":"Parameter-Efficient Quantum NLP for Paraphrase Detection: Performance, Robustness, and Entanglement","domain":"quant-ph","source":"arxiv","authors":["Farha Nausheen","Khandakar Ahmed","Farina Riaz"],"reviews":0,"subject":"quant-ph","summary":"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…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-09-13","reactions":{}}