Photonics has the potential to transform high-speed and energy-efficient artificial intelligence. However, most photonic computing platforms still rely on costly electro-optic conversions for data input, weight loading and storage, resulting in substantial overhead during large-scale computation. Here, we present a large-scale photonic in-memory computing platform based on liquid crystal (LC)-enabled Poincaré-sphere-connected neural networks. Megabyte-level nonvolatile photonic memory is introduced through light-driven reorientation of sulfonic azo-dye and LC molecules, which also functions as the reconfigurable optical neurons. This architecture uniquely integrates optical sensing, memory, and computing in situ, allowing minimal data-movement overheads in various switchable machine vision tasks. Ultrahigh scalability is demonstrated by a 9.44-million-neuron photonic processor array that supports massively parallel optical computing, recording and retrieval, providing a memory capacity 100,000 times larger than state-of-the-art photonic in-memory computing. This work offers a novel pathway toward scalable submicrometer-precision photonic memory, thus establishing a new paradigm for physical neural networks and brain-like artificial intelligence.
Abstract A research team demonstrated a 2067-km field-deployed optical frequency dissemination link over standard telecommunications fiber. The system achieved a record fractional frequency instability of 2.9 × 10⁻ 21 at one day averaging time, and maintained continuous phase lock for more than four days, marking a major step toward large-scale optical clock networks and future precision timing infrastructure.
Abstract Real-time intelligent systems increasingly require hardware that can adapt continuously to evolving inputs, yet most existing processors rely on static-weight inference, making them vulnerable to distribution shifts and error accumulation in dynamic environments. Although adaptive weight updates can, in principle, address this limitation, their implementation on electronic hardware is hindered by the stability–plasticity trade-off, as well as by the memory wall and clocking bottlenecks that become particularly severe in sequential processing. Here, we present a Temporally Plastic Photonic Processor (TPPP) that enables ultra-fast in situ adaptation by combining multi-timescale photonic kernels with a recursive optical delay memory. The architecture integrates a slow, reconfigurable kernel for stable long-term processing and a fast, dynamic kernel for transient adaptation, enabling time-varying weights to be embedded directly in the photonic domain without repeated electronic memory access. We experimentally validate the TPPP on linear and nonlinear sequential tasks. In both regimes, data-driven temporal plasticity enables the TPPP to outperform conventional static photonic baselines in robustness and accuracy. Under an operation-matched INT8 comparison, scaling analysis projects up to 16 × higher per-operation energy efficiency and up to 100 × lower intrinsic single-pass compute delay than advanced electronic processors, establishing the TPPP as a promising hardware framework for real-time adaptive photonic computing.
Abstract Hybrid metasurface–refractive systems have often raised concerns about their overall volume and the complexity of mechanical alignment. This reported study presents a fully integrated meta-aspheric optical system that realizes a wide field of view within an ultracompact volume of 0.02 cm3.
Optical skyrmions, particle-like topological textures of light with resilience to perturbation, hold great promise for next-generation robust information carriers. Recent advances have enabled their efficient generation and modulation in both classical and quantum regimes via artificial nanostructures with certain topological landscapes. However, existing studies are largely confined to (quasi)monochromatic domain, because the resonance-based linear and nonlinear light-matter interactions are constrained by narrowband response and strong spectral dispersion. Therefore, the exploration of coloured and white light skyrmions is virtually zero, blocking their extension to broadband information technologies. Here, we present an ultra-compact micro-generator that generates ultra-broadband coloured skyrmions from a natural ferroelectric spherulite crystal through the combined action of the photonic spin-orbital coupling and optical focusing. This approach circumvents the resonance effects intrinsic to artificially nanostructured optical systems. The resulting polychromatic skyrmions cover the entire visible spectral range and can propagate over an appreciable distance in free space. Their topological textures can be continually modulated by tuning the polarization of the incident beam, enabling switching among multiple topologies, such as skyrmions, biskyrmions, and quadrumerons. Furthermore, we reported the experimental evidence of spontaneous parametric down conversion process occurring within the ferroelectric spherulite, which implies the possibility to generate the correlated topological quantum states in further research. These distinctive features will revolutionize modern informatic applications spanning optical communication, data storage, and topological photonic devices.
