QECSync's decoder and compiler design draws on published quantum error correction literature and original technical work. We share our benchmark methodology and support independent verification.
We demonstrate that incorporating device-specific gate error rates and T1/T2 times into the MWPM edge weight matrix yields substantial improvements in logical error suppression compared to a uniform-weight baseline. Results are reported for synthetic noise models calibrated from three commercial superconducting processors at code distances d=5 and d=7.
Read on arXivA description of our benchmarking methodology for Union-Find decoder latency and accuracy measurements. Includes noise model definitions, hardware parameter ranges, and the open-source simulation harness used to generate all reported numbers. Intended to enable independent replication.
Request PDF via contact formAnalysis of factory count and qubit overhead for T-gate distillation as a function of physical error rate and code distance. We show that compiler-aware factory scheduling reduces total qubit overhead by 18–24% compared to static factory count estimates, across a range of circuit T-counts from 10 to 10,000 T-gates.
Read on arXivPresentation of early QECSync architecture decisions, the DeviceSpec abstraction layer design, and initial benchmark data comparing hardware-tuned vs hardware-agnostic decoder configurations.
Request slides via contact formQECSync's decoder implementation is grounded in the body of work on topological quantum codes and efficient decoding algorithms, particularly the surface code and its variants.
Key references: Fowler et al. "Surface codes: Towards practical large-scale quantum computation" (2012); Dennis et al. "Topological quantum memory" (2002); Delfosse & Nickerson "Almost-linear time decoding algorithm for topological codes" (2021); Higgott "PyMatching" (2022); Riesebos et al. Union-Find decoder complexity analysis.
Surface codes: Towards practical large-scale quantum computation
Physical Review A 86, 032324Topological quantum memory
J. Math. Phys. 43, 4452Almost-linear time decoding algorithm for topological codes
Quantum 5, 595PyMatching: A Python package for decoding quantum codes
ACM Transactions on Quantum Computing 3, 1If you are working on quantum error correction research and are interested in collaboration, access to benchmark data, or detailed discussion of our methodology, reach out.
Contact Research Team