TheMachine Press

A daily newspaper for the age of artificial intelligence.

Morning editionPermanent story

research

A Larger Surface Code Need Not Take Longer to Decode

A parallel sparse-blossom construction proves the same correction as the serial decoder with subconstant average time per syndrome round.

Published Updated Story ID: mp-2026-09-14-015
Read the complete editionStory JSON

Summary

A parallel sparse-blossom construction proves the same correction as the serial decoder with subconstant average time per syndrome round.

For rotated surface codes below a finite physical-error threshold, the proposed parallel sparse-blossom framework exactly matches the original decoder’s correction. Its average runtime over O(d) extraction rounds is quasi-polylogarithmic in code distance d, making average runtime per round o(1) for a d-round window. Simulations identify regimes where per-round time falls as distance rises. The result is an asymptotic algorithmic guarantee plus numerical study, not a hardware latency measurement.

Why it matters

A parallel sparse-blossom construction proves the same correction as the serial decoder with subconstant average time per syndrome round.

Limits and context

  • The result is an asymptotic algorithmic guarantee plus numerical study, not a hardware latency measurement.

Key claims

  1. A parallel sparse-blossom construction proves the same correction as the serial decoder with subconstant average time per syndrome round.

    Qualification: The result is an asymptotic algorithmic guarantee plus numerical study, not a hardware latency measurement.

    Evidence: source-2026-09-14-017

Sources

  1. arXiv preprint 2609.12262arXiv · primary research

Corrections

No corrections have been recorded for this story.