SearcharxivSearch

arXiv subjects

Shuming Zhang

Publications and source records attributed to Shuming Zhang.

5 recordsLinked to original sources

Fractality-induced photonic topological insulators

Fractal lattices have recently emerged as a promising setting for topological wave physics, but in most realizations the topological character is inherited from externally engineered couplings, gauge fields, or temporal modulation rather than from the fractal geometry itself. Here, we experimentally realize a photonic higher-order topological insulator in which the topology is induced solely by the self-similar geometry of a Sierpiński-gasket lattice. Following the isospectral reduction method recently proposed by Eek \textit{et al.}~\cite{Eek2025}, we show that the fractal waveguide array with uniform nearest-neighbor couplings can be mapped onto an effective breathing Kagome model that supports corner states. We selectively excite these modes with a weakly coupled detuned auxiliary waveguide and directly observe robust corner localization in real space, whereas an otherwise equivalent uniform triangular lattice exhibits only bulk diffraction under the same protocol. Spectral analysis and open-boundary calculations associate the observed states with nontrivial $C_3$ rotational topology, and disorder measurements further show that the corner localization persists over a finite range of random and symmetry-preserving disorder. Our results establish fractal geometry itself as a mechanism for generating topological boundary states in photonic lattices.

physics.optics

Nonlinear Photonic Tripartite Phase

Anderson localization is usually understood as a transition between extended and localized phases, with criticality confined to a single mobility edge. Recent advances predict that quasiperiodic systems can instead host a finite critical window bounded by mobility edges, in which localized, critical and extended states coexist. Yet both the experimental realization of this regime and whether interactions can provide controlled access to it remain unknown. Here, we realize such a tripartite phase in a nonlinear quasiperiodic photonic lattice and show that Kerr nonlinearity, acting as an effective interaction, enables state-selective access to the critical window. By tracking wavepacket dynamics, we distinguish localized, critical and extended transport regimes and uncover a state-selective response: rather than simply reinforcing localization through self-trapping, weak nonlinearity drives low-energy localized states into the critical window, whereas stronger nonlinearity restores localization. By contrast, critical, extended and high-energy localized states evolve monotonically towards self-trapped behaviour. Our results reveal a state-selective mechanism by which interactions provide controlled access to a pre-existing critical window in quasiperiodic systems.

cond-mat.mes-hall

Is AI Ready for Multimodal Hate Speech Detection? A Comprehensive Dataset and Benchmark Evaluation

Hate speech online targets individuals or groups based on identity attributes and spreads rapidly, posing serious social risks. Memes, which combine images and text, have emerged as a nuanced vehicle for disseminating hate speech, often relying on cultural knowledge for interpretation. However, existing multimodal hate speech datasets suffer from coarse-grained labeling and a lack of integration with surrounding discourse, leading to imprecise and incomplete assessments. To bridge this gap, we propose an agentic annotation framework that coordinates seven specialized agents to generate hierarchical labels and rationales. Based on this framework, we construct M^3 (Multi-platform, Multi-lingual, and Multimodal Meme), a dataset of 2,455 memes collected from X, 4chan, and Weibo, featuring fine-grained hate labels and human-verified rationales. Benchmarking state-of-the-art Multimodal Large Language Models reveals that these models struggle to effectively utilize surrounding post context, which often fails to improve or even degrades detection performance. Our finding highlights the challenges these models face in reasoning over memes embedded in real-world discourse and underscores the need for a context-aware multimodal architecture. Our dataset and code are available at https://github.com/mira-ai-lab/M3.

cs.MA

High-order Discontinuity Detection Physics-Informed Neural Network

In order to solve the problem of the difficult direct measurement of temperature field in fluid machinery under high-speed compressible conditions, this study combines high-order finite difference numerical format, Weighted Essentially Non-Oscillatory (WENO) discontinuity detection, and traditional Physics-Informed Neural Network (PINN) to develop a high-order discontinuity detection PINN (Hodd-PINN) that can achieve temperature field inversion with a small number of measurement points. When dealing with pure convection problems, Hodd-PINN introduces a 7th-order discretization for the convection term, reducing an additional 9.7% error compared to traditional low-order discretization methods. When dealing with pure diffusion problems, Hodd-PINN introduces an 8th-order discretization for the diffusion term, reducing an additional 12.8% error compared to traditional low-order discretization methods. In addition, this paper develops a loss function based on WENO discontinuity detection technology, which helps eliminate false discontinuities, allowing Hodd-PINN to successfully identify sparse waves that are easily overlooked in PINN's predicted results, reducing the error by 24.2%. Through extensive testing, this paper points out that the Hodd-PINN, which incorporates high-order discretization and discontinuity detection technology, can further reduce the prediction error of PINN, effectively reducing the data requirement, and can effectively solve the problem of false discontinuities. This method has important value for the inversion of temperature and velocity fields in fluid machinery under high-speed compressible conditions.

physics.flu-dyn

Construction method for general phenomenological RANS turbulence model

This paper proposes a phenomenological Reynolds Averaged Navier-Stokes (RANS) calculation model based on physical constraints. In this model part of the source terms in the e equation was replaced with the deep learning model, using the standard k-e model as a template. The simulation results of this model achieved a high error reduction of 51.7 % compared to the standard k-e model. To improve the adaptability and accuracy compared to the convergence of the abnormal flow regime, the coordinate technology proposed in this study was used in the modelling process. For the training data, the k-field and e-field were automatically corrected using this approach when the flow state deviated from the theoretical assumption. Based on the coordinate technology, a deep learning model for the source term of the equation was built, and the simulation error was reduced by 6.2 % compared to the uncoordinated one. From the results, the proposed coordinate technology can effectively be adapted to the underdeveloped flow state and assist in the more accurately modelling of the phenomenological RANS calculation model under a complex flow state.

physics.flu-dyn