Xian-Min Jin

The spontaneous breaking of the time translation symmetry has led to the emergence of quantum time crystals as a novel phase of matter, which is a phenomenon sustained by quantum coherence over long periods. Among them, discrete time crystals are recognized by the spontaneous generation of robust subharmonic oscillations. So far, most time crystals are restricted to exhibit a “period‐doubled” behavior. However, a fertile ground in time crystalline with period‐‐tupling has been suggested. Here we experimentally realize scalable period‐2‐tupling and period‐‐tupling in the one‐ (1D) and two‐dimensional (2D) photonic lattices, respectively. In this non‐interacting system, no spontaneous symmetry breaking occurs; instead, the period‐‐tupling arises from the preparation of specific superpositions of eigenmodes with quasienergy gaps of . All levels of the 2D spectrum are paired with different gaps, so that different response periods can be achieved by exciting different sites of the same lattice. The period ‐tupling is robust against perturbations over an extraordinarily long time evolution. The results may promote further research into arbitrary higher‐order time crystals in interacting photonic systems and into non‐equilibrium physics adopting Floquet driving. Next‐generation quantum devices, such as topologically protected quantum state registers, quantum sensors, and simulators, can also be inspired by this approach.

The photon systems have exhibited remarkable performance in the field of quantum science, especially in quantum communication and quantum computing. This achievement is largely attributed to the quantum correlations created by optical nonlinear processes in certain materials. However, a trade-off between the brightness and correlation quality of photon-pair sources imposes a fundamental limit on the maximum achievable number of photons in large-scale quantum-advantage experiments. Besides, there still exists a lack of complete experimental evidence exploring how the bunching property of driving light affects the light-matter interaction mechanism, thereby limiting the understanding of multi-body dynamics. Here, we experimentally implement, to our knowledge, the first observation of the photon correlation inheritance effect (CIE) at an extremely low noise level. We observed that the correlation of driving light is linearly transferred to the quantum correlation of excited photons with high fidelity within a photon-atom hybrid dynamics process. We verified that utilizing this inheritance effect allows for the linear manipulation of photon excited rate while preserving its non-classical nature, thereby circumventing the fundamental trade-off in nonlinear excitation processes. Based on this, we demonstrate that CIE can multiplicatively enhance the performance of quantum emitters and quantum multi-body dynamics sensors. Our research fills a gap in experimental phenomena within the domain of nonlinear light-matter interaction and provides novel perspectives and avenues for future research in optical quantum science.

Photonic neural network chips promise compact footprint, low latency, and high energy efficiency. Yet, their scale and computing throughput are fundamentally constrained by one-dimensional input interfaces, unavoidable waveguide crossings, and the resulting crosstalk and excess loss. As a result, two-dimensional (2D) image data must be serialized through limited input ports, sacrificing spatial parallelism and creating input/output (I/O) bottlenecks. Here we demonstrate a programmable three-dimensional (3D) photonic neural network chip, fabricated by femtosecond laser direct writing (FLDW) in glass, that directly processes 2D images. The cascaded architecture alternates photonic-lantern waveguide arrays and phase-shifter arrays to implement matrix operations. An 8-layer 8 × 8 device achieves a computing throughput of 6554 TOPS, surpasses leading planar photonic platforms, and delivers 93% accuracy on MNIST classification and 94% fidelity in optical pattern generation. By combining 3D spatial parallelism with programmability, this work establishes a scalable paradigm for reconfigurable photonic computing in complex inference tasks.

The de Broglie–Bohm (dBB) theory was first proposed by L. de Broglie in 1927, nearly a century ago, and has now become an indispensable complement to the orthodox interpretation of quantum theory. As a nonlocal and contextual hidden‐variable theory, the dBB theory is compatible with Bell's inequality and the Kochen–Specker theorem. With the establishment and development of the weak measurement formalism, average Bohmian trajectories were experimentally reconstructed in 2011. Subsequent experiments in 2016 and 2019 provide a visual manifestation of the predicted nonlocality of Bohmian trajectories. Recently, stemming from the Englert–Scully–Süssmann–Walther (ESSW) Gedanken experiment, the concept of “nonlocal energy alteration” (i.e., nonlocal change of energy in space) has been proposed and observed on a quantum memory platform. This paper first reviews the evolution of the dBB theory, then summarizes recent experimental research progress, particularly providing insights into the redistribution of energy in space‐like intervals, and finally prospects for future research in this field.

In-memory computing, which enables computation directly within memory, represents an efficient approach to processing massively parallel computation tasks that are intractable for conventional computers. However, implementations of in-memory computing have been primarily limited to the classical regime, with its nonclassical counterpart yet to be fully explored. Quantum memory, with its unique capability to generate, preserve, and nontrivially operate on quantum states, offers spectacular quantum-enhanced advantages and is thus a promising candidate for in-memory computing. Here, leveraging a room-temperature quantum memory, we demonstrate a quantum-enhanced and reconfigurable in-memory stochastic computing system, where correlated photons, randomly produced in the quantum memory, serve as the computing resources. We show that addition and multiplication operations can be straightforwardly achieved by accumulating photon counts, and multiple computing tasks can be accelerated by mapping them into parallel accumulations of photon counts. Furthermore, the calculation results are obtained through stochastic processes, ensuring security in remote computation since no efficient information can be distinguished by eavesdropping on a small portion of the computation data. This in-memory computing system is enhanced by nonclassical correlations, which accelerate computing process and may stimulate future research and applications in the emerging field of quantum-enhanced computing architectures.

Integrated photonic computing possesses remarkable advantages in computational power, becoming a promising candidate for high-performance computing in the post-Moore era. However, integrated photonic computing systems face trade-offs between scale, speed, and precision, hindering their development into high-performance processors. In this work, we propose and experimentally demonstrate a photonic quantized neural network using a large-scale lithium niobate on insulator high-speed electro-optic engine. With the electro-optic engine providing 1.6 TeraOPS computational power, our photonic quantized neural network experiments achieve a high accuracy of 97% in CNN-based MNIST handwritten digit recognition and 80% in GCN-based Cora data node classification, while also realizing over 95% model compression. Based on the quantized input-weight parallel encoding scheme, our high-speed electro-optic engine achieves a 100-fold improvement in system frequency over digital computing architectures and a 10-fold improvement in both modulation speed and compute density over optoelectronic computing architectures, while also achieving competitive energy efficiency. Our experimental findings indicate that our photonic quantized neural network technology paves the way for large-scale and high-speed integrated photonic computing, further promoting its practical application.

Twisted light with orbital angular momentum (OAM) introduces an innovative dimension in modern optics, providing a valuable resource for advancing both classical and quantum information technologies. The utilization of OAM and multi-core multiplexing in advanced optical fiber communication systems has been demonstrated as the pivotal solution to the escalating demand for information transmission in contemporary society. However, creating a large-scale and stable device for spatially separated cores supporting OAM modes is more challenging compared to established technologies for spectrum and polarization, which have mature devices. To address this challenge, we propose and fabricate a double-layer doughnut waveguide supporting three OAM modes under 1550 nm, using femtosecond laser direct writing technology. Based on that, we design the 37-core integrated space-division multiplexing device with uniform performance by depth-related power compensation. Our platform represents a breakthrough by overcoming the current technological barriers in on-chip multi-mode support and large-scale uniformity. This advancement enables a theoretical increase over a hundred fold in single-fiber information capacity, positioning it well ahead of existing integrated solutions. Designed for next-generation communication systems, it optimizes integrated space-division multiplexing, setting new benchmarks for efficiency and scalability in optical networks and facilitating unprecedented advancements in communication technologies.

In a study of amplifying atomic spin waves, we observe a nontrivial phenomenon: The spin wave stored in moving atoms has a capability of absorbing energy from an external light source, and exhibits a regeneration process. We demonstrate that this regeneration significantly enhances the lifetime and retrieval efficiency of the spin wave, while concurrently the noise is effectively suppressed. Our results suggest that the regeneration mechanism holds promise for mitigating the pronounced decoherence typically encountered in spin waves carried by room-temperature media, therefore offering potential applications in the realms of quantum information and precision measurements at ambient conditions.

Quantum tunnelling, a hallmark phenomenon of quantum mechanics, allows particles to pass through the classically forbidden region. It underpins fundamental processes ranging from nuclear fusion and photosynthesis to the operation of superconducting qubits. Yet the underlying dynamics of particle motion during tunnelling remain subtle and are still the subject of active debate. Here, by analyzing the temporal evolution of the tunnelling process, we show that the particle velocity inside the barrier continuously relaxes from a large initial value toward a smaller one, and may even approach zero in the evanescent regime. Meanwhile, the probability density within the barrier gradually builds up before reaching its stationary profile, in contrast to existing inherently. In addition, starting from the steady-state equations, we derive an explicit relation between the particle velocity and the barrier width, and show that the velocity in evanescent states approaches zero when the barrier is sufficiently wide. These findings resolve the apparent paradox of a vanishing steady-state velocity coexisting with a finite particle density. We point out that defining an effective speed from the probability density, rather than from the probability current, can lead to spuriously nonzero "stationary speed," as appears to be the case in Ref. [Nature 643, 67 (2025)]. Our work establishes a clear dynamical picture for the formation of tunnelling flow and provides a theoretical foundation for testing time-resolved tunnelling phenomena.