SearcharxivSearch

arXiv subjects

Xinxing Zhou

Publications and source records attributed to Xinxing Zhou.

At least 19 recordsLinked to original sources

Omnidirectional photonic chiral flatband in nonlocal membrane metasurfaces

Omnidirectional flat-band resonances, characterized by an enhanced photonic density of states and inherent angular robustness, are highly sought-after in integrated nanophotonic devices, particularly when integrated with chiral functionality. Here we realize such resonances in a nonlocal silicon membrane metasurface patterned with periodic square-lattice air-hole arrays. Increasing the lattice period not only compresses the Brillouin zone but, crucially, weakens the evanescent coupling between neighbouring Bloch modes associated with the same-order guided resonances. Driven by the tight-binding model in the limit of weak inter-unit-cell coupling, the pronounced band flattening of the degenerate guided resonance along both $k_{x}$ and $k_{y}$ yields, giving rise to an omnidirectional flat-band resonance. Remarkably, both numerical simulations and experiments reveal a universal route for endowing flat-band guided resonances with optical chirality through the deliberate breaking of the mirror symmetry of air holes. As a result, the omnidirectional chiral flat-band resonance emerges along both principal in-plane directions, with $Q$-factors exceeding 10$^{3}$ and circular dichroism greater than 0.9 over a wide angular range of $\pm 5^{\circ}$. Nonlinear measurements further show that the resulting resonance not only drives highly efficient third-harmonic generation but also imparts a pronounced spin-selective character to the nonlinear process. Simultaneously, the highly efficient nonlinear process also enables chirality-controlled frequency-upconversion imaging. Our results establish a general paradigm for engineering omnidirectional chiral flat-band resonances in planar silicon platforms, opening new opportunities for nonlinear nanophotonics and chiral imaging.

physics.optics

EIDOS: Latent-Space Predictive Learning for Time Series Foundation Models

Most time series foundation models are pretrained by directly predicting future observations, which often yields weakly structured latent representations that capture surface noise rather than coherent and predictable temporal dynamics. In this work, we introduce EIDOS, a foundation model family that shifts pretraining from future value prediction to latent-space predictive learning. We train a causal Transformer to predict the evolution of latent representations, encouraging the emergence of structured and temporally coherent latent states. To ensure stable targets for latent-space learning, we design a lightweight aggregation branch to construct target representations. EIDOS is optimized via a joint objective that integrates latent-space alignment, observational grounding to anchor representations to the input signal, and direct forecasting supervision. On the GIFT-Eval benchmark, EIDOS mitigates structural fragmentation in the representation space and achieves state-of-the-art performance. These results demonstrate that constraining models to learn predictable latent dynamics is a principled step toward more robust and reliable time series foundation models.

cs.LG

Photonic spin Hall effect dependent on Landau level transitions in monolayer WTe2

Landau level (LL) engineered photonic spin Hall effect (PSHE) holds great promise for nanoscale manipulation and steering of magneto-optical transport in two-dimensional atomic systems. Herein, we theoretically investigate PSHE modulated by LL transitions {\delta}n = n'-n =-2, 0, +2 (where n and n' indicate the LL indexes of valence and conduction bands, respectively) in monolayer WTe2. Results show that PSHE tuned by {\delta}n =-2, 0, +2 has completely different dependent behaviors on LLs, incident angle of incident photons, and magnetic induction intensity. These discrepancies are ascribed to Hall-conductivity-incurred Hall angle {\Theta} because the variation tendency of photonic spin Hall shifts is similar to that of {\Theta} with changing the LL index. Giant PSHE with the largest in-plane displacement of more than 400 times of incident wavelength is obtained at the transition |n=55>->|n'=57>. Remarkably enhanced PSHE occurs at near-zero Hall angles. In-plane and transverse spin-dependent displacements give their respective extremum values at the same incident angles when the {\Theta} is near to zero, and their incident-angle deviation will become larger and larger as the |{\Theta}| increases. This unambiguously confirms the strong influence of Hall angle in the PSHE, shedding important insights into the fundamental properties of spin-orbit interaction of light in time-reversal symmetry breaking quantum systems.

cond-mat.mes-hall

Loss-free enhancement of photonic spin Hall shift by electromagnetically induced transparency

The photonic spin Hall effect (PSHE), a result of spin-orbit interaction, has attracted significant interest because of its fundamental importance and potential applications. Optical losses are ubiquitous, which inherently suppress the photonic spin Hall shift (PSHS). In this work, we consider an atomic medium that exhibits both absorption and transparency to investigate and mitigate the effects of loss on PSHS. We demonstrate that laser-induced coherence in an atomic medium, leading to electromagnetically induced transparency (EIT) at resonance, counteracts the detrimental effects of losses on the PSHS. Upon EIT in a coherent medium enclosed within dielectric slabs, the reflectivity of the incident polarized state is reduced near Brewster's angle to enhance PSHS. Moreover, the tunable refractive index of the atomic medium enables the manipulation of PSHS without structural modifications with a tiny loss. Our proposed loss-free approach to PSHS may enable advanced optical sensing and other spin-based applications.

physics.optics

EffiCANet: Efficient Time Series Forecasting with Convolutional Attention

The exponential growth of multivariate time series data from sensor networks in domains like industrial monitoring and smart cities requires efficient and accurate forecasting models. Current deep learning methods often fail to adequately capture long-range dependencies and complex inter-variable relationships, especially under real-time processing constraints. These limitations arise as many models are optimized for either short-term forecasting with limited receptive fields or long-term accuracy at the cost of efficiency. Additionally, dynamic and intricate interactions between variables in real-world data further complicate modeling efforts. To address these limitations, we propose EffiCANet, an Efficient Convolutional Attention Network designed to enhance forecasting accuracy while maintaining computational efficiency. EffiCANet integrates three key components: (1) a Temporal Large-kernel Decomposed Convolution (TLDC) module that captures long-term temporal dependencies while reducing computational overhead; (2) an Inter-Variable Group Convolution (IVGC) module that captures complex and evolving relationships among variables; and (3) a Global Temporal-Variable Attention (GTVA) mechanism that prioritizes critical temporal and inter-variable features. Extensive evaluations across nine benchmark datasets show that EffiCANet achieves the maximum reduction of 10.02% in MAE over state-of-the-art models, while cutting computational costs by 26.2% relative to conventional large-kernel convolution methods, thanks to its efficient decomposition strategy.

cs.LG

Towards Universal Large-Scale Foundational Model for Natural Gas Demand Forecasting

In the context of global energy strategy, accurate natural gas demand forecasting is crucial for ensuring efficient resource allocation and operational planning. Traditional forecasting methods struggle to cope with the growing complexity and variability of gas consumption patterns across diverse industries and commercial sectors. To address these challenges, we propose the first foundation model specifically tailored for natural gas demand forecasting. Foundation models, known for their ability to generalize across tasks and datasets, offer a robust solution to the limitations of traditional methods, such as the need for separate models for different customer segments and their limited generalization capabilities. Our approach leverages contrastive learning to improve prediction accuracy in real-world scenarios, particularly by tackling issues such as noise in historical consumption data and the potential misclassification of similar data samples, which can lead to degradation in the quaility of the representation and thus the accuracy of downstream forecasting tasks. By integrating advanced noise filtering techniques within the contrastive learning framework, our model enhances the quality of learned representations, leading to more accurate predictions. Furthermore, the model undergoes industry-specific fine-tuning during pretraining, enabling it to better capture the unique characteristics of gas consumption across various sectors. We conducted extensive experiments using a large-scale dataset from ENN Group, which includes data from over 10,000 industrial, commercial, and welfare-related customers across multiple regions. Our model outperformed existing state-of-the-art methods, demonstrating a relative improvement in MSE by 3.68\% and in MASE by 6.15\% compared to the best available model.

cs.LG

Breakdown of effective-medium theory by a photonic spin Hall effect

Effective-medium theory pertains to the theoretical modelling of homogenization, which aims to replace an inhomogeneous structure of subwavelength-scale constituents with a homogeneous effective medium. The effective-medium theory is fundamental to various realms, including electromagnetics and material science, since it can largely decrease the complexity in the exploration of light-matter interactions by providing simple acceptable approximation. Generally, the effective-medium theory is thought to be applicable to any all-dielectric system with deep-subwavelength constituents, under the condition that the effective medium does not have a critical angle, at which the total internal reflection occurs. Here we reveal a fundamental breakdown of the effective-medium theory that can be applied in very general conditions: showing it for deep-subwavelength all-dielectric multilayers even without critical angle. Our finding relies on an exotic photonic spin Hall effect, which is shown to be ultra-sensitive to the stacking order of deep-subwavelength dielectric layers, since the spin-orbit interaction of light is dependent on slight phase accumulations during the wave propagation. Our results indicate that the photonic spin Hall effect could provide a promising and powerful tool for measuring structural defects for all-dielectric systems even in the extreme nanometer scale.

physics.optics

Anomalous spatial shifts in interface electronic reflection beyond linear approximation

Recently, the electronic analogy of the anomalous spatial shift, including Goos-H\"{a}nchen and Imbert-Fedorov effects, has been attracting widespread interest. The current research on the anomalous spatial shift in interface electronic reflection is based on the paradigm of linear approximation, under which the center position of the incident and reflected beams are obtained by expanding the phases of relevant basis states and scattering amplitudes to the first order of incident momentum. However, in a class of normal cases, the linear approximation can lead to a divergent spatial shift in reflection for certain incident angles even though the corresponding reflection possibility is finite. In this work, we show that such non-physical results are caused by an abrupt change in the number of the propagating states at critical parameters, and can be resolved by calculating the center positions of the scattering beams beyond the linear approximation. Moreover, we find that the beam width has an important influence on the spatial shift near the critical angles. We demonstrate our idea via concrete calculations of Goos-H\"{a}nchen and Imbert-Fedorov shift on two representative models. These results are beneficial for clarifying the scope of application of the linear approximation in the study of anomalous spatial shifts.

cond-mat.mes-hall

Wide-angle giant photonic spin Hall effect

Photonic spin Hall effect is a manifestation of spin-orbit interaction of light and can be measured by a transverse shift \lambda of photons with opposite spins. The precise measurement of transverse shifts can enable many spin-related applications, such as precise metrology and optical sensing. However, this transverse shift is generally small (i.e. \delta /\lambda <{10}^{-1}, \lambda is the wavelength), which impedes its precise measurement. To-date proposals to generate giant spin Hall effect (namely with \delta /\lambda >{10}^{2}) have severe limitations, particularly its occurrence only over a narrow angular cone (with a width of \Delta \theta <{1}^{\circ}). Here we propose a universal scheme to realize the wide-angle giant photonic spin Hall effect with \Delta \theta >{70}^{\circ} by exploiting the interface between free space and uniaxial epsilon-near-zero media. The underlying mechanism is ascribed to the almost-perfect polarization splitting between s and p polarized waves at the designed interface. Remarkably, this almost-perfect polarization splitting does not resort to the interference effect and is insensitive to the incident angle, which then gives rise to the wide-angle giant photonic spin Hall effect.

physics.optics

Controlling photonic spin Hall effect via exceptional points

The photonic spin Hall effect (SHE), featured by a spin-dependent transverse shift of an impinging optical beam driven by its polarization handedness, has many applications including precise metrology and spin-based nanophotonic devices. It is highly desirable to control and enhance the photonic SHE. However, such a goal remains elusive, due to the weak spin-orbit interaction of light, especially for systems with optical loss. Here we reveal a flexible way to modulate the photonic SHE via exceptional points, by exploiting the transverse shift in a parity-time (PT) symmetric system with balanced gain and loss. The underlying physics is associated with the near-zero value and abrupt phase jump of the reflection coefficients at exceptional points. We find that the transverse shift is zero at exceptional points, but it is largely enhanced in their vicinity. In addition, the transverse shift switches its sign across the exceptional point, resulting from spontaneous PT-symmetry breaking. Due to the sensitivity of transverse shift at exceptional points, our work also indicates that the photonic SHE can enable a precise way to probe the location of exceptional point in photonic systems.

physics.optics

Spin photonics and spin-photonic devices with dielectric metasurfaces

Dielectric metasurfaces with spatially varying birefringence and high transmission efficiency can exhibit exceptional abilities for controlling the photonic spin states. We present here some of our works on spin photonics and spin-photonic devices with metasurfaces. We develop a hybrid-order Poincare sphere to describe the evolution of spin states of wave propagation in the metasurface. Both the Berry curvature and the Pancharatnam-Berry phase on the hybrid-order Poincare sphere are demonstrated to be proportional to the variation of total angular momentum. Based on the spin-dependent property of Pancharatnam-Berry phase, we find that the photonic spin Hall effect can be observed when breaking the rotational symmetry of metasurfaces. Moreover, we show that the dielectric metasurfaces can provide great flexibility in the design of novel spin-photonic devices such as spin filter and spin-dependent beam splitter.

physics.optics

Modified weak measurements for detecting photonic spin Hall effect

Weak measurement is an important technique for detecting the tiny spin-dependent splitting in photonic spin Hall effect. The weak measurement is only valid when the probe wavefunction remains almost undisturbed during the procedure of measurements. However, it does not always satisfy such condition in some practical situations, such as in the strong-coupling regime or the preselected and postselected states are nearly orthogonal. In this paper, we develop a modified weak measurement for detecting photonic spin Hall effect when the probe wavefunction is distorted. We find that the measuring procedure with preselected and postselected ensembles is still effective. This scheme is important for us to detect the photonic spin Hall effect in the case where neither weak nor strong measurements can detect the spin-dependent splitting. The modified theory is valid not only in weak-coupling regime but also in the strong-coupling regime, and especially in the intermediate regime. The theoretical models of conventional weak measurements and modified weak measurements are established and compared. We show that the experimental results coincide well with the predictions of the modified theory.

physics.optics

Hybrid-order Poincaré sphere

In this work, we develop a hybrid-order Poincaré sphere to describe the evolution of polarization states of wave propagation in inhomogeneous anisotropic media. We extend the orbital Poincaré sphere and high-order Poincaré sphere to a more general form. Polarization evolution in inhomogeneous anisotropic media with special geometry can be conveniently described by state evolution along the longitude line on the hybrid-order Poincaré sphere. Similar to that in previously proposed Poincaré spheres, the Berry curvature can be regarded as an effective magnetic field with monopole centered at the origin of sphere and Berry connection can be interpreted as the vector potential. Both the Berry curvature and the Pancharatnam-Berry phase on the hybrid-order Poincaré sphere are demonstrated to be proportional to the total angular momentum. Our scheme provides a convenient method to describe the spin-orbit interaction in inhomogeneous anisotropic media.

physics.optics

Realization of tunable spin-dependent splitting in intrinsic photonic spin Hall effect

We report the realization of tunable spin-dependent splitting in intrinsic photonic spin Hall effect. By breaking the rotational symmetry of a cylindrical vector beam, the intrinsic vortex phases that the two spin components of the vector beam carries, which is similar to the geometric Pancharatnam-Berry phase, is no longer continuous in the azimuthal direction, and leads to observation of spin accumulation at the opposite edge of the beam. Due to the inherent nature of the phase and independency of light-matter interaction, the observed photonic spin Hall effect is intrinsic. Modulating the topological charge of the vector beam, the spin-dependent splitting can be enhanced and the direction of spin accumulation is switchable. Our findings may provide a possible route for generation and manipulation of spin-polarized photons, and enables spin-based photonics applications.

physics.optics

Photonic spin Hall effect for precision metrology

The photonic spin Hall effect (SHE) is generally believed to be a result of an effective spin-orbit coupling, which describes the mutual influence of the spin (polarization) and the trajectory of the light beam. The photonic SHE holds great potential for precision metrology owing to the fact that the spin-dependent splitting in photonic SHE are sensitive to the physical parameter variations of different systems. Remarkably, using the weak measurements, this tiny spin-dependent shifts can be detected with the desirable accuracy so that the corresponding physical parameters can be determined. Here, we will review some of our works on using photonic SHE for precision metrology, such as measuring the thickness of nanometal film, identifying the graphene layers, detecting the strength of axion coupling in topological insulators, and determining the magneto-optical constant of magnetic film.

physics.optics

Photonic spin Hall effect in metasurfaces with rotational symmetry-breaking

Observation of photonic spin Hall effect (SHE) in dielectric-based metasurfaces with rotational symmetry breaking is presented. We find that the spin-dependent splitting is a unique angular splitting in the real position space, and is attributed to the space-variant Pancharatnam-Berry phase (PB). Breaking the rotational symmetry of the PB phase by misalignment of the central axes of the incident beam and the metasurface, the spin-dependent shift is observable. We show that the spin-dependent shift can be enhanced by increasing the rotation rate of the metasurface, so the metasurface provides a great flexibility in the manipulation of photonic SHE.

physics.optics

Generation of cylindrical vector vortex beams by two cascaded metasurfaces

We present a simple and efficient method to generate any cylindrical vector vortex (CVV) beams based on two cascaded metasurfaces. The metasurface works as a space-variant Panchratnam-Berry phase element and can produce any desirable vortex phase and vector polarization. The first metasurface is used to switch the sign of topological charges associated with vortex, and the second metasurface is applied to manipulate the local polarization. This method allows us to simultaneously manipulate polarization and phase of the CVV beams.

physics.optics

Generation of arbitrary cylindrical vector beams on the higher order Poincare sphere

We propose and experimentally demonstrate a novel interferometric approach to generate arbitrary cylindrical vector beams on the higher order Poincare sphere. Our scheme is implemented by collinear superposition of two orthogonal circular polarizations with opposite topological charges. By modifying the amplitude and phase factors of the two beams, respectively, any desired vector beams on the higher order Poincare sphere with high tunability can be acquired. Our research provides a convenient way to evolve the polarization states in any path on the high order Poincare sphere.

physics.optics