Logical qubits already outperform physical ones on a real task

Pauline Mathiot, Elio Garnaoui, Axel-Ugo Leriche, Evan Philip, Boris Albrecht, Clémence Briosne-Fréjaville, Lorenzo Cardarelli, Antoine Cornillot, Gwennolé Cournez, Luc Couturier, Julius De Hond, Rebecca El Koussaifi, Thomas Eritzpokoff, Florian Fasola, Antonio Andrea Gentile, Casper Gyurik, Clotilde Hamot, Loïc Henriet, Gaétan Hercé, Michael Kaicher, Lucas Lassablière, François-Marie Le Régent, Edgar Leroux, Yohann Machu, Hadriel Mamann, Luis Ortiz, Annie Paine, Thomas Pansiot, Arnaud Peloquin, Francisco Ponciano, Julien Ripoll, Raja Selvarajan, Adrien Signoles, Henrique Silvério, Siddhy Tan, Marie Taouzinet, Selim Touati, Louis Vignoli, Antoine Browaeys, Pascal Scholl

Running a machine-learning algorithm for solving differential equations on the same neutral-atom hardware at both the physical and logical qubit levels, the logical implementation produced more accurate quantum kernels despite using more quantum resources. The improvement was traced directly to noise errors that the error-correcting encoding caught and suppressed. It's an early but concrete demonstration that fault-tolerant overhead already pays off on end-to-end tasks, not just in abstract error benchmarks.