19 January 2024 in Science
To harness the potential of a quantum computer, quantum information must be protected against error by encoding it into a logical state that is suitable for quantum error correction. The Gottesman-Kitaev-Preskill (GKP) qubit is a promising candidate because the required multiqubit operations are readily available at optical frequency. To date, however, GKP qubits have been demonstrated only at mechanical and microwave frequencies. We realized a GKP state in propagating light at telecommunication wavelength and verified it through homodyne measurements without loss corrections. The generation is based on interference of cat states, followed by homodyne measurements. Our final states exhibit nonclassicality and non-Gaussianity, including the trident shape of faint instances of GKP states. Improvements toward brighter, multipeaked GKP qubits will be the basis for quantum computation with light.
29 January 2025 in Nature Photonics
Entanglement is a fundamental resource for various optical quantum information processing (QIP) applications. To achieve high-speed QIP systems, entanglement should be encoded in short wavepackets. Here we report the real-time observation of ultrafast optical Einstein–Podolsky–Rosen correlation at a picosecond timescale in a continuous-wave system. Optical phase-sensitive amplification using a 6-THz-bandwidth waveguide-based optical parametric amplifier enhances the effective efficiency of 70-GHz-bandwidth homodyne detectors, mainly used in 5G telecommunication, enabling its use in real-time quantum state measurement. Although power measurement using frequency scanning, such as an optical spectrum analyser, is not performed in real time, our observation is demonstrated through the real-time amplitude measurement and can be directly used in QIP applications. The observed Einstein–Podolsky–Rosen states show quantum correlation of 4.5 dB below the shot-noise level encoded in wavepackets with 40 ps period, equivalent to 25 GHz repetition—103 times faster than previous entanglement observation in continuous-wave systems. The quantum correlation of 4.5 dB is already sufficient for several QIP applications, and our system can be readily extended to large-scale entanglement. Moreover, our scheme has high compatibility with optical communication technology such as wavelength-division multiplexing, and femtosecond-timescale observation is also feasible. Our demonstration is a paradigm shift in accelerating accessible quantum correlation—the foundational resource of all quantum applications—from the nanosecond to picosecond timescales, enabling ultrafast optical QIP.
19 August 2024 in Physical Review A
Non-Gaussian quantum gates are essential components for optical quantum information processing. However, the efficient implementation of practically important multimode higher-order non-Gaussian gates has not been comprehensively studied. We propose a measurement-based method to directly implement general, multimode, and higher-order non-Gaussian gates using only fixed non-Gaussian ancillary states and adaptive linear optics. Compared to existing methods, our method allows for a more resource-efficient and experimentally feasible implementation of multimode gates that are important for various applications in optical quantum technology, such as the two-mode cubic quantum nondemolition gate or the three-mode continuous-variable Toffoli gate, and their higher-order extensions. Our results will expedite the progress toward fault-tolerant universal quantum computing with light. Published by the American Physical Society 2024
13 May 2026 in Physical Review X
Advanced quantum technologies rely on non-Gaussian states of light, essential for universal quantum computation, fault-tolerant error correction, and quantum sensing. Their practical realization, however, faces hurdles: Simulating large multimode generators is computationally demanding, and benchmarks such as the do not capture how effectively photon detections yield useful non-Gaussianity. We address these challenges by introducing the ( s 0 , δ 0 ) , a continuous and operational measure that goes beyond stellar rank. Leveraging these parameters, we develop a universal optimization method that reduces photon-number requirements and greatly enhances success probabilities while preserving state quality. Applied to the Gottesman-Kitaev-Preskill state generation, for example, our method cuts the required photon detections by a factor of 3 and raises the preparation probability by nearly 10 8 . Demonstrations across cat states, cubic phase states, Gottesman-Kitaev-Preskill states, and even random states confirm broad gains in experimental feasibility. Our results provide a unifying principle for resource-efficient non-Gaussian state generation, charting a practical route toward scalable optical quantum technologies and fault-tolerant quantum computation.
12 August 2025 in Physical Review Research
Continuous-variable (CV) quantum information processing is a promising candidate for large-scale fault-tolerant quantum computation. However, analysis of CV quantum process relies mostly on direct computation of the evolution of operators in the Heisenberg picture, and the features of CV space has yet to be thoroughly investigated in an intuitive manner. One key ingredient for further exploration of CV quantum computing is the construction of a computational model that brings visual intuition and new tools for the analysis of quantum protocols, especially non-Gaussian processes. In this paper, we delve into a graphical computational model, inspired by a similar model for qubit-based systems called the ZX calculus, which enables the representation of arbitrary CV quantum process as a simple directed graph. We demonstrate the utility of our model as a graphical tool to comprehend CV processes intuitively by showing how equivalences between two distinct quantum processes can be proven as a sequence of diagrammatic transformations for various quantum protocols. We also examine possible applications of our model, such as measurement-based quantum computing, characterization of Gaussian and non-Gaussian processes, and circuit optimization.