Back to News
quantum-computing

Record Fidelity Achieved for Entangled Logical Qubits on IBM Hardware

Quantum Computing Report
Loading...
2 min read
0 likes
⚡ Quantum Brief
A multinational research team achieved record 98% fidelity for entangled logical qubits on IBM’s 127-qubit superconducting processor, surpassing prior transmon-based benchmarks of 79–93%. The breakthrough combines Quantum Error Detection with Normalizer Dynamical Decoupling (NDD), a novel technique repurposing logical code elements to suppress both physical and logical errors without added qubits. NDD operates at the logical level, avoiding the control overhead of physical decoupling while mitigating ZZ crosstalk—a persistent bottleneck in quantum scaling. Experiments showed logical Bell states sustained 91–94% fidelity over 55 microseconds, marking "beyond-breakeven" performance compared to unencoded qubits. Published in Nature Communications, the study demonstrates a hardware-efficient path to fault tolerance, addressing error accumulation without increasing code distance.
Record Fidelity Achieved for Entangled Logical Qubits on IBM Hardware

Summarize this article with:

Record Fidelity Achieved for Entangled Logical Qubits on IBM Hardware Researchers from Quantum Elements, USC’s Center for Quantum Information Science & Technology, IBM, and RWTH Aachen University have demonstrated a significant milestone in fault-tolerant quantum computing by achieving record-high fidelities for entangled logical qubits. Published in Nature Communications, the study utilized a 127-qubit IBM superconducting processor to implement a novel hybrid protocol combining Quantum Error Detection (QED) with a technique termed Normalizer Dynamical Decoupling (NDD). This approach addresses a critical bottleneck in quantum scaling: the accumulation of logical errors, particularly ZZ crosstalk, which typically bypasses standard physical-level decoupling and fixed-distance error correction codes. The technical breakthrough centers on the use of NDD, where the normalizer elements of the quantum code itself—specifically the [[4, 2, 2]] code—are repurposed as dynamical decoupling pulses. Unlike physical DD, which operates on individual qubits and often introduces control overhead that scales with the system size, NDD operates at the logical level. By using logical-level operations as pulses, the researchers suppressed both logical and physical error channels simultaneously. This hybrid strategy is hardware-efficient because it uses a fixed set of logical pulse generators, effectively extending the performance of the code without requiring an increase in code distance or additional physical qubits. Experimental results confirmed “beyond-breakeven” performance, with logical Bell states maintaining significantly higher fidelity over time compared to the best physical (unencoded) pairs on the same hardware. Key metrics from the study include: Peak Post-selected Fidelity: 98% for encoded Bell-state preparation, a substantial increase over previous transmon-based benchmarks (typically 79–93%). Long-Duration Stability: Average logical Bell state fidelities were maintained between 91–94% over a 55 μs window. Error Suppression: The protocol identified and mitigated logical Z errors arising from crosstalk, which were determined to be the dominant error source in the 127-qubit transmon architecture. For the complete technical results and methodology, consult the Nature Communications paper here. March 4, 2026 Mohamed Abdel-Kareem2026-03-04T15:40:46-08:00 Leave A Comment Cancel replyComment Type in the text displayed above Δ This site uses Akismet to reduce spam. Learn how your comment data is processed.

Read Original

Tags

superconducting-qubits
quantum-computing
quantum-hardware
quantum-error-correction

Source Information

Source: Quantum Computing Report