2025/04/25 by Jean-Baptiste Waring, Waring, Jean-Baptiste, Christophe Père +3
Computer Science · #FOS: Physical sciences #Quantum Computing Algorithms and Architecture #Quantum Information and Cryptography #Quantum Physics (quant-ph) #Quantum-Dot Cellular Automata
paper · pdf · doi:10.48550/arxiv.2504.18429
openalex publication_date 2025/04/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Scalable quantum computing relies on high-quality, long-range entanglement, a challenge on noisy, near-term devices. The need for practical insights for near-term algorithm design calls for trade-offs exploration in implementing dynamic circuits on current hardware. In this work, we experimentally compare three CNOT implementations for generating Bell states across varying qubit separations on a 127-qubit IBM Quantum Eagle processor (ibmquebec): a unitary (SWAP-based) approach, a dynamic approach with mid-circuit measurements and classical feedforward, and a post-processed approach. We use Clauser-Horne-Shimony-Holt (CHSH) inequality violations to quantify entanglement quality. We observe that, beyond 10 qubits, dynamic circuits lead to higher |S| values than the unitary approach, demonstrating improved distance-dependent entanglement preservation. The post-processed approach yields the highest CHSH values, reaching |S| > 2 up to 13 qubits. Our results underscore the critical need for faster classical feedforward and higher readout fidelity.