Integrated multimode quantum optics is a promising platform for scalable continuous-variable quantum technologies leveraging multimode squeezing in both the spatial and spectral domains. However, on-chip measurement, routing and processing the relevant “supermodes” over which the squeezing resource is distributed still scales quadratically with the number of modes N , causing a rapid increase in photonic circuit size and number of required measurements. Here, we introduce a variational scheme, relying on self-configuring photonic networks (SCNs) that learn and extract the most-squeezed supermodes sequentially, reducing both the circuit size and the experimental overhead. Using homodyne measurement as a cost function, a sparse SCN discovers the l ≪ N most significant supermodes using O ( l N ) physical elements and optimization steps. We analyze and numerically simulate these architectures for both real-space and frequency-domain implementations, showing a fidelity close to unity between the learned circuit and the supermode decomposition, even in the presence of optical losses and detection noise. In the frequency domain, we show that circuit size can be further reduced by using inverse-designed surrogate networks, which emulate the layers learned thus far. Using two different frequency encoding schemes—uniformly and nonuniformly spaced frequency bins—we reduce an entire network (learning all N supermodes) to O ( N ) and even O ( 1 ) modulated cavities. Our results point toward chip-scale, resource-efficient quantum processing units and demultiplexers for continuous variable processing in multimode quantum optics, with applications ranging from quantum communication, metrology, and computation.
Integrated multimode quantum optics is a promising platform for scalable continuous-variable quantum technologies leveraging multimode squeezing in both the spatial and spectral domains. However, on-chip measurement, routing and processing the relevant “supermodes” over which the squeezing resource is distributed still scales quadratically with the number of modes N , causing a rapid increase in photonic circuit size and number of required measurements. Here, we introduce a variational scheme, relying on self-configuring photonic networks (SCNs) that learn and extract the most-squeezed supermodes sequentially, reducing both the circuit size and the experimental overhead. Using homodyne measurement as a cost function, a sparse SCN discovers the l ≪ N most significant supermodes using O ( l N ) physical elements and optimization steps. We analyze and numerically simulate these architectures for both real-space and frequency-domain implementations, showing a fidelity close to unity between the learned circuit and the supermode decomposition, even in the presence of optical losses and detection noise. In the frequency domain, we show that circuit size can be further reduced by using inverse-designed surrogate networks, which emulate the layers learned thus far. Using two different frequency encoding schemes—uniformly and nonuniformly spaced frequency bins—we reduce an entire network (learning all N supermodes) to O ( N ) and even O ( 1 ) modulated cavities. Our results point toward chip-scale, resource-efficient quantum processing units and demultiplexers for continuous variable processing in multimode quantum optics, with applications ranging from quantum communication, metrology, and computation.
Multimode squeezed light is a key resource for high-dimensional quantum technologies, enhancing metrological sensitivity, boosting communication security, and enabling parallel processing in computation. Its practical potential, however, remains constrained by the inherent single-mode operation of homodyne detection, necessitating post-processing for multimode characterization. Here, we overcome this long-standing challenge by employing multimode optical parametric amplification, enabling loss-tolerant direct detection of squeezing in each mode, which in turn permits mode sorting after amplification. As a result, we demonstrate, for the first time to the best of our knowledge, the real-time monitoring of multimode squeezing. With a spatial light modulator sorting the modes, we simultaneously measure squeezing in nine spatial modes co-propagating within one beam. Although mode sorting and filtering reduce the detection efficiency to less than 0.3%, we observe high-purity squeezing of up to - 7.9 ± 0.6 dB - to the best of our knowledge, the highest squeezing recorded for pulsed light. Furthermore, we demonstrate real-time, loss-tolerant characterization of continuous-variable entanglement and extend it to the detection of cluster states. Similar methods can be applied in the frequency domain, facilitating a crucial capability for scalable quantum technologies.
Scalable interferometers lie at the heart of photonic quantum technologies, but their expansion has been fundamentally limited by optical losses that grow with circuit depth. Here, we introduce and experimentally demonstrate a measurement-induced architecture for multimode squeezed-light interferometers that overcomes this barrier. By shifting complexity from deep optical networks to programmable homodyne measurements, we realize effective transformations within a shallow, low-loss platform. We validate the principle with a six-mode device and extend it to a 400-mode interferometer, marking a leap in scale beyond conventional designs. Crucially, this strategy not only enables scalable squeezed light interferometry but also provides a powerful route to the generation of large-scale entangled states-a key requirement for quantum computing, simulation, and communication. Our results establish measurement-induced circuits as a practical pathway toward noisy intermediate-scale quantum (NISQ) applications and future demonstrations of quantum advantage.
Continuous-variable (CV) multipartite entangled states, particularly Greenberger–Horne–Zeilinger (GHZ) states, are an indispensable resource for quantum networks and quantum information processing. Conventional approaches to generating CV GHZ states rely on combining multiple independently prepared squeezed lights through complex interferometric networks, which impose significant challenges in scalability and stability. Here, we experimentally demonstrate a compact and resource-efficient scheme for generating a tripartite CV GHZ state from a single quantum frequency comb produced by a synchronously pumped optical parametric oscillator. By employing multipixel homodyne detection, we simultaneously measure quadrature fluctuations across multiple frequency bands and reconstruct the full multimode covariance matrix in a single measurement. Through eigenmode decomposition, we extract a set of independent squeezed supermodes, which are subsequently transformed via a tailored linear unitary operation implemented in post-processing. Using this approach, we deterministically construct a tripartite GHZ state from three supermodes and verify its entanglement via the van Loock–Furusawa criteria, with the corresponding correlation variances measured to be 0.57, 0.75, and 0.57, all below the shot noise limit. This method effectively realizes a scalable entanglement generation scheme within a single optical beam, eliminating the need for complex interferometric networks. Our results provide a scalable and versatile platform for multimode quantum information processing.
Squeezed-light generation in photonic integrated circuits (PICs) is essential for scalable continuous-variable (CV) quantum information processing. By suppressing quantum fluctuations below the shot-noise limit, squeezed states enable quantum-enhanced sensing and serve as a standard resource for CV quantum information processing. While chip-level squeezed-light sources have been demonstrated, extending this capability to the wafer level with reproducible strong squeezing to bolster large-scale quantum-enhanced sensing and information processing has been hindered by squeezed light's extreme susceptibility to device imperfections. Here, we report wafer-scale fabrication, generation, and characterization of two-mode squeezed-vacuum states on a fully complementary metal-oxide-semiconductor (CMOS)-compatible silicon nitride (Si3N4) PIC platform. Across a 4-inch wafer, 8 dies yield 2.9-3.1 dB directly measured quadrature squeezing with <0.2 dB variation, demonstrating excellent uniformity. This performance is enabled by co-integrating ultralow-loss, strongly overcoupled high-Q microresonators, cascaded pump-rejection filters, and low-loss inverse-tapered edge couplers. The measurements agree with a first-principles theoretical model parameterized solely by independently extracted device parameters and experimental settings. The measured squeezing level can be further improved by enhancing the efficiencies of off-chip detection and chip-to-fiber coupling. These results establish a reproducible, wafer-scale route to nonclassical-light generation in integrated photonics and lay the groundwork for scalable CV processors, multiplexed entanglement sources, and quantum-enhanced sensing.
Squeezed light underpins quantum-enhanced sensing and continuous-variable quantum information processing, and integrated photonics offers a route to producing it at scale. Universal to these applications are squeezed-light generation and measurement. Importantly, quantum measurements serve not only as readout but also as active operations in quantum-state evolution. However, integrating squeezed-light generation and photodetection on the same photonic chip has remained challenging because they impose fundamentally conflicting material requirements: low optical loss to preserve quantum correlations, but efficient photon absorption for photodetection. Here, we demonstrate squeezed-light generation, routing, and balanced homodyne detection integrated on a single photonic chip through heterogeneous integration. A two-mode squeezed quantum microcomb comprising 34 quantum modes is measured with approximately 3 dB squeezing. Our work establishes a scalable architecture for fully integrated squeezed-light quantum photonic systems, unifying quantum-state generation, processing, and detection on a single chip.
We introduce a new class of self-configuring photonic architectures called variational optical processors (VOPs). These devices tackle the problem of decomposing and measuring multimode light fields, both partially coherent and quantum, without requiring prior knowledge of the modes involved. Classical strategies for modal decomposition—such as Karhunen-Loève expansions [1] for partially coherent beams or Schmidt decompositions [2] for bipartite quantum states—often rely on comprehensive tomography and complex data processing. By contrast, VOPs discover the relevant modes in situ through a simple optimization of detection signals at their outputs (e.g., measured optical power or coincidence counts), eliminating the overhead usually associated with scanning and reconstruction. Prior demonstrations of self-configuring photonic networks have proven the feasibility of such approaches for analyzing and generating multimode coherent optical fields [3].