A new class of error-correction algorithms called phantom codes could significantly improve the accuracy of quantum computations, according to research led by Shayan Majidy at Harvard University. Computer simulations show the approach delivers results up to 100 times more accurate than conventional error-correction programs on certain tasks.
Errors remain one of the most persistent obstacles in quantum computing. While existing error-correction methods can protect stored information reliably, they tend to falter during active computation, particularly in programs that involve many sequential steps.
What Makes Phantom Codes Different
Quantum computers process information using physical units called qubits. For error-resistant computation, devices typically group qubits into logical qubits and then manipulate them physically, by firing lasers or microwaves, to entangle them or alter their quantum properties. Each such physical action is a potential source of error.
Phantom codes sidestep this requirement. Rather than generating entanglement through physical manipulation, they exploit entanglement that already exists within the computation. Because the entanglement is already present, no additional physical action is needed to create it. Fewer interventions means fewer opportunities for errors to enter the system.
“It’s not a free lunch. It’s just a lunch that was already there and we weren’t eating it,” Majidy said.
His team tested the approach in simulations on two specific tasks: preparing a qubit state commonly used in quantum computations, and modeling a quantum material. In both cases, phantom codes required fewer physical manipulations and produced substantially more accurate results than standard methods.
Scope and Limitations
Phantom codes are not a universal fix. Majidy acknowledges they perform best in computations that already require significant entanglement, and offer little advantage in programs that do not.
Mark Howard at the University of Galway offered a practical analogy, comparing the choice of error-correction code to selecting armor. Plate armor offers strong protection but is heavy and restrictive; chain mail is more flexible but less protective. Phantom codes behave like chain mail, offering flexibility at the cost of requiring more qubits than some traditional approaches. Howard suggests they may prove useful for specific subroutines within larger quantum programs, but are unlikely to resolve the error problem across the board.
Dominic Williamson at the University of Sydney notes that how well phantom codes compete against other error-correction methods remains an open question, one that may partly depend on how quantum hardware develops.
What Comes Next
Majidy’s team is already working with researchers who build quantum computers from ultracold atoms, the architecture used by QuEra. He expects the insights from phantom codes to feed into a broader shift in how quantum programs are designed, moving toward implementations that are tailored to specific tasks and the practical capabilities of particular hardware.
The research is available on arXiv under DOI: 10.48550/arXiv.2601.20927.
Photo by Andrey Matveev on Unsplash
This article is a curated summary based on third-party sources. Source: Read the original article
