YLYi Liang

Yi Liang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yi Liang has a total of 178 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Anomalous Deterministic Localization Driven by Staggered Disorder

    by Peilong Hong, Yu Wang, Fuhao Ji, Mingfang Yi, Liwei Zhang, Peng Cheng, Wangping Cheng, Yi Liang, Guoquan Zhang

    Disorder scattering occurs broadly in wave‐matter systems, offering vast potential for applications ranging from random lasing to quantum information processing. However, achieving customized control over wave transport and localization in disordered systems remains a significant challenge. Notably, wave transport in the weak scattering regime typically exhibits energy‐dilute diffusion, while in the strong scattering regime, Anderson localization is inherently stochastic and unpredictable in location. Here we report an anomalous phase termed staggered‐disorder induced localization (SDIL) in disordered structure, which achieves robust and deterministic localization in disordered systems. Specifically, the SDIL halts wave expansion along a predefined direction while facilitating orthogonal transport within a 2D structure. The SDIL is found persisting across a wide range of disorder strengths for a finite 2D structure. Remarkably, the transport trajectory is fully configurable by engineering the staggered disorder, providing an attractive mechanism for wave routing in disordered systems. Our quasi‐normal mode analysis reveals a localized phase region, wherein emergent high‐quality modes drive this robust localized scattering. These findings establish SDIL as a missing paradigm in scattering physics, providing a universal framework for achieving deterministic wave localization and routing within disordered systems.

  • All‐Optical Nonlinear Activation Functions Based on Photonic Crystal Microcavity

    by Zitao Wei, Zi'ANg Guo, Wei Wu, Qinglian Li, Lun Qu, Lin Li, Xiaohai Liu, Yi Liang, Mengxin Ren, Jingjun Xu

    The growing computational demand of artificial intelligence (AI) calls for energy‐efficient computing architectures. Diffractive deep neural networks () offer a promising route for optical computing, but their lack of nonlinear activation functions limits their representation capability. Here, we demonstrate a passive all‐optical Leak‐ReLU nonlinear activation function based on a 1D photonic crystal microcavity. By exploiting resonantly enhanced thermo‐optic nonlinearity, the cavity exhibits a temperature‐induced resonance blueshift that produces intensity‐dependent nonlinear transmission with tunable activation thresholds through wavelength selection. Integrating the experimentally characterized nonlinear activation function into a improves the classification accuracy by 1.79% on Fashion‐MNIST compared with a two‐layer linear and accelerates convergence by 2.6 times on MNIST. This work provides a compact and passive approach for implementing nonlinear optical neural networks and advances low‐power neuromorphic optical computing.

  • Tunable rotational transverse multifoci generated by catastrophe fields with discrete phase and coordinate rotation

    by Yao-Xi Chu, Si-Yan Liao, Nan Wang, Xiao Liang, Xiang-Gao Wang, Yu-Xuan Ren, Yi Liang

  • Interplay of flat-band and Anderson localization in disordered moiré superlattices

    by Qian Liu, Xiaoshuang Xia, Junjie Wang, Peilong Hong, Lei Xu, Lujun Huang, Daohong Song, Yi Liang

    Disorder in moir\'e superlattices simultaneously degrades flat-band localization and induces Anderson localization, yet how these two regimes interact has remained unclear. Here, we introduce a combined framework linking localization\penalty1000-\hskip0ptlength scaling with differential probability density analysis to map localization transitions in partially disordered one\penalty1000-\hskip0ptdimensional silicon moir\'e lattices. It is found that flat bands confined within the interband gap keep their strong localization even as disorder grows. In contrast, flat bands intersecting dispersive bands exhibit rich behaviors: the low\penalty1000-\hskip0ptfrequency branch undergoes an inverse Anderson transition, while the high\penalty1000-\hskip0ptfrequency branch supports coexisting flat-band and Anderson localization at strong disorder. Our results deliver direct evidence of competing localization mechanisms in disordered moir\'e systems and offer guiding principles for engineering robust, nonideal moir\'e photonic devices.

  • Open Access

    Tunable rotational transverse multifoci generated by catastrophe fields with discrete phase and coordinate rotation

    by Yaoxi Chu, Siyan Liao, Nan Wang, Xiao Liang, Xianggao Wang, Yu-Xuan Ren, Yi Liang

    Autofocusing structured fields have important potential in parallel optical manipulation and configurable focal field. However, it is still challenging to regulate the transverse multifoci array, as superimposed catastrophe fields tend to concentrate energy near the center or mainly exhibit longitudinal autofocusing. Here, we propose a transverse multifocal higher-order catastrophe field by engineering discrete phase on the constituent subfields. This strategy enables coplanar rotational multifocal arrays with tunable radius, focal spacing, focal width, and residual central intensity through the subfield number N and structural parameter C. We show that the transverse focusing pattern is governed by discrete phase symmetry: removal of the phase collapses the array into a single central focus, whereas prescribed discrete phase delays produce well-separated rotational arrays of foci. The design theory has been well corroborated by experiment using off-axis computer-generated holography. Phase, transverse energy-flow, and orbital angular momentum (OAM) analysis reveals orientation evolution, circulating transverse energy flow, and nonzero multimodal OAM without an externally imposed vortex phase. The established results suggest that the discrete subfield phase engineering can be considered as a universal strategy for tailoring rotational multifoci catastrophe fields, opening opportunities for structured-light engineering, parallel optical manipulation, and topological light-field design.

  • Nonmonotonic Q -factor evolution of flatband modes in disordered finite moiré photonic crystals

    by Qian Liu, Junjie Wang, Luyao Zhang, Xiang Yao, Xianggao Wang, Peilong Hong, Yi Liang

    This study investigates radiative loss within a finite, single-cell 3D moir\'e photonic crystal---the fundamental building block for practical moir\'e devices. The authors demonstrate a counterintuitive, nonmonotonic dependence of the average quality factor on disorder strength. Specifically, they reveal that weak disorder actively suppresses radiative coupling to enhance the quality factor, whereas stronger disorder exacerbates scattering into the radiation light cone, severely degrading optical confinement.

  • Evolution of flatband localization in disordered Moiré Lieb superlattice

    by Jia He, Haoming Chen, Qian Liu, Xiaoshuang Xia, Mingtao Ding, Yunyun Xu, Junjie Wang, Xiaoming Mo, Peilong Hong, Yi Liang


  • Universal transverse confinement and scaling of transmission eigenchannels in disordered photonic lattices

    by Peilong Hong, Wenjing Li, Xiangqi Huang, Rong Tao, Yu Wang, Mingfang Yi, Liwei Zhang, Peng Cheng, Wangping Cheng, Yuandi He, et al.

    Harnessing subsets of scattering paths, particularly through the transmission eigenchannels, has recently become a crucial strategy to control wave transport in disordered media. While previous works are mainly focused on structures with both axial and horizontal disorders, the disordered photonic lattice—offering the unique advantage that its uniform axial direction acts as an effective time dimension—has remained largely unexplored. Here, we unveil the distinctive transverse profiles of transmission eigenchannels in disordered photonic lattices, which are inherently related to the intriguing long- and short-lived scattering modes by taking into account the effective temporal dimension. Interestingly, we demonstrate a remarkable localization-to-delocalization transition, i.e., high-transmission eigenchannels exhibit a strongly shrinking bell-shaped transverse profile, while low-transmission eigenchannels feature an extended profile with a characteristic central dip. Further theoretical analysis leads to a profound linear scaling between the transverse size and the transmission coefficient. Moreover, we find that the disorder strength can amplify the interference of scattering paths, leading to enhanced confinement for high-transmission channels but deepened central dip for the low-transmission counterparts. These results offer new perspectives on wave localization and energy flow in complex systems, with direct implications for enhancing performance in random lasers, photoluminescence, and nonlinear optical processes.

  • High-fidelity image reconstruction from dynamic nonlinear speckles mediated by chlorophyll solution using deep learning

    by Lu Tian, Si-Yan Liao, Yao-Xi Chu, Xiao-Ying Tang, Shun-Yu Liu, Yu-Xuan Ren, Pei-Long Hong, Yi Liang

    The information recovery of speckles is crucial in applications such as underwater exploration, satellite remote sensing, and biomedical imaging. The interplay of nonlinearity and scattering breaks the wavefronts and destroys the image formation. Previous studies have primarily focused on linear scattering models, leaving the reconstruction of images distorted by nonlinear perturbations largely unexplored. Here, we employ a convolutional neural network named the Attention U-Net to reconstruct images modulated by a chlorophyll solution, exhibiting self-defocusing nonlinearity. The proposed approach exhibits strong generalization capability for unseen object categories under nonlinear perturbation. Furthermore, by comparing the performance of models trained with single-power and multi-power mixed datasets, we reveal the relationship between reconstruction ability and nonlinear optical response. Our study establishes an effective connection between green-material-induced nonlinear modulation and deep learning-based optical decoding, offering promising potential for applications in optical information encryption and intelligent imaging systems.