Why do quantum computers make so many errors?
Qubits are fragile: noise, heat and even cosmic rays can disturb them.
Ask your own in the app →Quantum computers use qubits, which are powerful but error-prone. The papers we have covered tackle errors from several sides: fixing results after a run, shielding chips from cosmic rays, searching for better error correction, and testing early uses like forecasting pollution. Most results are simulations or small devices, so progress is real but early.
Paper by paper
arXiv9 min listen
Quantum circuit design usually keeps one winner and discards the rest. This preprint mixes diverse quantum and classical models and beats the best single one.
Preprint Benchmark study Preliminary evidence
arXiv9 min listen
Quantum error correction can be tuned to a chip's noise at no hardware cost. This study builds a search that finds good tweaks far faster, and a new one.
Preprint Simulation study Preliminary evidence
arXiv8 min listen
Cosmic rays can knock out several qubits at once. A few-atom-thick boron nitride layer might block them. This underground study checks it does no harm first.
Preprint Device measurements Preliminary evidence
arXiv9 min listen
A quantum neural network on a 172-qubit emulator forecast air pollution well, but with wavelet-cleaned data a small classical network nearly matched it.
Preprint Emulated quantum ML benchmark Preliminary evidence
arXiv8 min listen
Terahertz light is a gap for quantum technology. This preprint builds an artificial molecule from two tiny graphene islands with levels about 0.145 THz apart.
Preprint Device experiment + modelling Preliminary evidence
arXiv9 min listen
Error-corrected quantum computers still let some mistakes through. A Google team proposes cleaning up the answers afterwards, using alarms already recorded.
Preprint Theory + simulations Preliminary evidence
Questions people ask
Qubits are fragile: noise, heat and even cosmic rays can disturb them.
Ask your own in the app →Tuning how qubits are checked to a chip's noise can help without extra hardware, in simulation.
Ask your own in the app →Not clearly: in one forecasting study a small classical network nearly matched the quantum one.
Ask your own in the app →