arXiv subjects

Shu Chen

Publications and source records attributed to Shu Chen.

At least 55 records · Page 3Linked to original sources

arXiv2025

Two-dimensional talc as a natural hyperbolic material

This study demonstrates that two-dimensional talc, a naturally abundant mineral, supports hyperbolic phonon-polaritons (HPhPs) at mid-infrared wavelengths, thus offering a low-cost alternative to synthetic polaritonic materials. Using scattering scanning near-field optical microscopy (s-SNOM) and synchrotron infrared nano spectroscopy (SINS), we reveal tunable HPhP modes in talc flakes of a long lifetime. These results highlight the potential of natural 2D talc crystals to constituting an effective platform for establishing scalable optoelectronic and photonic devices.

cond-mat.mes-hall↗

arXiv2025

Types of dynamical behavior in a quasiperiodic mosaic lattice

Quasiperiodic mosaic systems with the quasiperiodic potential being added periodically with a fixed lattice interval have attracted significant attention due to their peculiar spectral properties with exactly known mobility edges, which separate localized from delocalized states. These mobility edges do not vanish even in the region of large quasiperiodic potential strength, although the width of the energy window of extended states decreases with the increase in potential strength and thus becomes very narrow in the limit of strong quasiperiodic disorder. In this paper, we study the dynamics of a quasiperiodic mosaic lattice and unravel its peculiar dynamical properties. By scrutinizing the expansion dynamics of wave packet and the evolution of density distribution, we unveil that the long-time density distribution displays obviously different behaviors at odd and even sites in the region of large quasiperiodic potential strength. Particularly, the timescale of dynamics exhibits an inverse relationship with the quasiperiodic potential strength. To understand these behaviors, we derive an effective Hamiltonian in the large quasiperiodic potential strength region, which is composed of decoupled Hamiltonians defined on the odd and even sites, respectively. While all eigenstates of the effective Hamiltonian defined on even sites are localized, the eigenstates of effective Hamiltonian defined on odd sites include both localized and extended eigenstates. Our results suggest that the effective Hamiltonian can describe the dynamical behaviors well in the large quasiperiodic potential strength region and provides an intuitive framework for understanding the peculiar dynamical behaviors in the quasiperiodic mosaic lattice.

cond-mat.dis-nn↗

arXiv2025

Lindbladian dynamics with loss of quantum jumps

The Lindblad master equation (LME) describing the Markovian dynamics of the quantum open system can be understood as the evolution of the effective non-Hermitian Hamiltonian balanced with random quantum jumps. Here we investigate the balance-breaking dynamics by partly eliminating jumps from postselection experiments. To describe this dynamics, a non-linear Lindblad master equation (NLME) is derived from quantum trajectory method. However, the NLME shows significant advantages in analytical analysis over quantum trajectory method. Using the NLME, we classify the dynamics into two classes. In the trivial class, the process of reducing jumps is completely equivalent to weakening the coupling from the environment. In contrast, the nontrivial class presents more complex dynamics. We study a prototypical model within this class and demonstrate the existence of the postselected skin effect whose steady state is characterized by the accumulation of particles on one side. The steady-state distribution can be fitted by a scale-invariant tanh function which is different from the uniform distribution of LME. Furthermore, the NLME can give a reasonable framework for studying the interplay and competition between the non-Hermitian Hamiltonians and dissipative terms. We show this by capturing the characteristics of the trajectory-averaged entanglement entropy influenced by non-Hermitian skin effect and Zeno effect in the model of postselected skin effect.

quant-ph↗

arXiv2024

Closed dynamical recursion equations for correlation functions and the application on the construction of Liouvillian spectrum in Lindbladian systems

For an open quantum system described by the Lindblad equation, full characterization of its dynamics typically needs the knowledge of the Liouvillian spectrum and correlation functions. Solving the Liouvillian spectrum and correlation functions are usually formidable tasks, and most previous studies are constrained to simple models and lower-order correlations. In this work, we derive a closed form of dynamical recursion equations for all even-order correlation functions associated with the quadratic Liouvillian superoperator, thus extending the Wick's theorem and enabling the reconstruction of the corresponding Liouvillian spectrum. Furthermore, we study the quartic Liouvillian superoperator, establishing the necessary and sufficient conditions for the closure of second-order correlation functions. Building on this, the dynamical expressions for even-order correlation functions under quadratic dissipation are derived, culminating in a method for constructing the Liouvillian spectrum in this extended framework.

quant-ph↗

arXiv2024

Production of open-charm pentaquark molecules in decay $B^0 \rightarrow \bar{D}^0 p \bar{p}$

This study investigates the production of open-charm pentaquark molecular states, specifically $N\bar{D}^*$ and $\bar{N}\bar{D}^*$, within the $B^0 \rightarrow \bar{D}^0 p \bar{p}$ decay. We examine the invariant mass spectra of $p\bar{D}^0$ and $\bar{p}\bar{D}^0$, incorporating the rescattering process through a quasipotential Bethe-Salpeter equation approach. Our findings suggest the possible identification of the isoscalar $\bar{N}\bar{D}^*$ molecule with $3/2^+$ as the antiparticle partner of $Λ_c(2940)$ in the $\bar{p}\bar{D}^0$ mass spectrum. Although a molecular state near the $\bar{N}\bar{D}$ threshold with $1/2^-$ exists, its signal is weak, indicating that the $B^0 \rightarrow \bar{D}^0 p \bar{p}$ decay may not be ideal for its detection. A distinct signal of the isovector $N\bar{D}^*$ molecule with $1/2^-$ may appear in the $p\bar{D}^0$ invariant mass spectrum, while the signal for the $3/2^-$ state remains very weak. We emphasize the importance of the three-body decay of the bottom meson as a valuable method for studying open-charm molecules and advocate for increased attention and more precise experimental measurements of the $B^0 \rightarrow \bar{D}^0 p \bar{p}$ decay.

hep-ph↗

arXiv2024

Universal Spreading Dynamics in Quasiperiodic Non-Hermitian Systems

Non-Hermitian systems exhibit a distinctive type of wave propagation, due to the intricate interplay of non-Hermiticity and disorder. Here, we investigate the spreading dynamics in the archetypal non-Hermitian Aubry-André model with quasiperiodic disorder. We uncover counter-intuitive transport behaviors: subdiffusion with a spreading exponent $δ=1/3$ in the localized regime and diffusion with $δ=1/2$ in the delocalized regime, in stark contrast to their Hermitian counterparts (halted vs. ballistic). We then establish a unified framework from random-variable perspective to determine the universal scaling relations in both regimes for generic disordered non-Hermitian systems. An efficient method is presented to extract the spreading exponents from Lyapunov exponents. The observed subdiffusive or diffusive transport in our model stems from Van Hove singularities at the tail of imaginary density of states, as corroborated by Lyapunov-exponent analysis.

cond-mat.dis-nn↗

arXiv2024

Scale-tailored localization and its observation in non-Hermitian electrical circuits

Anderson localization and non-Hermitian skin effect are two paradigmatic wave localization phenomena, resulting from wave interference and the intrinsic non-Hermitian point gap, respectively. In this study, we unveil a novel localization phenomenon associated with long-range asymmetric coupling, termed scale-tailored localization, where the number of induced localized modes and their localization lengths scale exclusively with the coupling range. We show that the long-range coupling fundamentally reshapes the energy spectra and eigenstates by creating multiple connected paths on the lattice. Furthermore, we present experimental observations of scale-tailored localization in non-Hermitian electrical circuits utilizing adjustable voltage followers and switches. The circuit admittance spectra possess separate point-shaped and loop-shaped components in the complex energy plane, corresponding respectively to skin modes and scale-tailored localized states. Our findings not only expand and deepen the understanding of peculiar effects induced by non-Hermiticity but also offer a feasible experimental platform for exploring and controlling wave localizations.

cond-mat.mes-hall↗

arXiv2024

Certifying Quantum Temporal Correlation via Randomized Measurements: Theory and Experiment

We consider the certification of temporal quantum correlations using the pseudo-density matrix (PDM), an extension of the density matrix to the time domain, where negative eigenvalues are key indicators of temporal correlations. Conventional methods for detecting these correlations rely on PDM tomography, which often involves excessive redundant information and requires exponential resources. In this work, we develop an efficient protocol for temporal correlation detection by virtually preparing the PDM within a single time slice and estimating its second-order moments using randomized measurements. Through sample complexity analysis, we demonstrate that our protocol requires only a constant number of measurement bases, making it particularly advantageous for systems utilizing ensemble average measurements, as it maintains constant runtime complexity regardless of the number of qubits. We experimentally validate our protocol on a nuclear magnetic resonance platform, a typical thermodynamic quantum system, where the experimental results closely align with theoretical predictions, confirming the effectiveness of our protocol.

quant-ph↗

arXiv2024

Statistical and dynamical aspects of quantum chaos in a kicked Bose-Hubbard dimer

Systems of interacting bosons in double-well potentials, modeled by two-site Bose-Hubbard models, are of significant theoretical and experimental interest and attracted intensive studies in contexts ranging from many-body physics and quantum dynamics to the onset of quantum chaos. In this work we systematically study a kicked two-site Bose-Hubbard model (Bose-Hubbard dimer) with the on-site potential difference being periodically modulated. Our model can be equivalently represented as a kicked Lipkin-Meshkov-Glick model and thus displays different dynamical behaviors from the kicked top model. By analyzing spectral statistics of Floquet operator, we unveil that the system undergoes a transition from regularity to chaos with increasing the interaction strength. Then based on semiclassical approximation and the analysis of Rényi entropy of coherent states in the basis of Floquet operator eigenstates, we reveal the local chaotic features of our model, which indicate the existence of integrable islands even in the deep chaotic regime. The semiclassical analysis also suggests that the system in chaotic regime may display different dynamical behavior depending on the choice of initial states. Finally, we demonstrate that dynamical signatures of chaos can be manifested by studying dynamical evolution of local operators and out of time order correlation function as well as the entanglement entropy. Our numerical results exhibit the richness of dynamics of the kicked Bose-Hubbard dimer in both regular and chaotic regimes as the initial states are chosen as coherent spin states located in different locations of phase space.

quant-ph↗

arXiv2024

Exact zeros of fidelity in finite-size systems as a signature for probing quantum phase transitions

The fidelity is widely used to detect quantum phase transitions, which is characterized by either a sharp change of fidelity or the divergence of fidelity susceptibility in the thermodynamical limit when the phase-driving parameter is across the transition point. In this work, we unveil that the occurrence of exact zeros of fidelity in finite-size systems can be applied to detect quantum phase transitions. In general, the fidelity $\mathcal{F}(γ,\tildeγ)$ always approaches zero in the thermodynamical limit, due to the Anderson orthogonality catastrophe, no matter whether the parameters of two ground states ($γ$ and $\tildeγ$) are in the same phase or different phases, and this makes it difficult to distinguish whether an exact zero of fidelity exists by finite-size analysis. To overcome the influence of orthogonality catastrophe, we study finite-size systems with twist boundary conditions, which can be introduced by applying a magnetic flux, and demonstrate that exact zeros of fidelity can be always accessed by tuning the magnetic flux when $γ$ and $\tildeγ$ belong to different phases. On the other hand, no exact zero of fidelity can be observed if $γ$ and $\tildeγ$ are in the same phase. We demonstrate the applicability of our theoretical scheme by studying concrete examples, including the Su-Schrieffer-Heeger model, Creutz model and Haldane model. Our work provides a practicable way to detect quantum phase transitions via the calculation of fidelity of finite-size systems.

quant-ph↗

arXiv2024

Spread complexity and dynamical transition in multimode Bose-Einstein condensates

We study the spread complexity in two-mode Bose-Einstein condensations and unveil that the long-time average of the spread complexity $\overline{C}_{K}$ can probe the dynamical transition from self-trapping to Josephson oscillation. When the parameter $ω$ increases over a critical value $ω_{c}$, we reveal that the spread complexity exhibits a sharp transition from lower to higher value, with the corresponding phase space trajectory changing from self-trapping to Josephson oscillation. Moreover, we scrutinize the eigen-spectrum and uncover the relation between the dynamical transition and the excited state quantum phase transition, which is characterized by the emergence of singularity in the density of states at critical energy $E_{c}$. In the thermodynamical limit, the cross point of $E_{c}(ω)$ and the initial energy $E_{0}(ω)$ determines the dynamical transition point $ω_{c}$. Furthermore, we show that the different dynamical behavior for the initial state at a fixed point can be distinguished by the long-time average of the spread complexity, when the fixed point changes from unstable to stable. Finally, we also examine the sensitivity of $\overline{C}_{K}$ for the triple-well bosonic model which exibits the transition from chaotic dynamics to regular dynamics.

cond-mat.quant-gas↗

arXiv2024

REInstruct: Building Instruction Data from Unlabeled Corpus

Manually annotating instruction data for large language models is difficult, costly, and hard to scale. Meanwhile, current automatic annotation methods typically rely on distilling synthetic data from proprietary LLMs, which not only limits the upper bound of the quality of the instruction data but also raises potential copyright issues. In this paper, we propose REInstruct, a simple and scalable method to automatically build instruction data from an unlabeled corpus without heavy reliance on proprietary LLMs and human annotation. Specifically, REInstruct first selects a subset of unlabeled texts that potentially contain well-structured helpful and insightful content and then generates instructions for these texts. To generate accurate and relevant responses for effective and robust training, REInstruct further proposes a rewriting-based approach to improve the quality of the generated instruction data. By training Llama-7b on a combination of 3k seed data and 32k synthetic data from REInstruct, fine-tuned model achieves a 65.41\% win rate on AlpacaEval leaderboard against text-davinci-003, outperforming other open-source, non-distilled instruction data construction methods. The code is publicly available at \url{https://github.com/cs32963/REInstruct}.

cs.CL↗

arXiv2024

Sample Enrichment via Temporary Operations on Subsequences for Sequential Recommendation

Sequential recommendation leverages interaction sequences to predict forthcoming user behaviors, crucial for crafting personalized recommendations. However, the true preferences of a user are inherently complex and high-dimensional, while the observed data is merely a simplified and low-dimensional projection of the rich preferences, which often leads to prevalent issues like data sparsity and inaccurate model training. To learn true preferences from the sparse data, most existing works endeavor to introduce some extra information or design some ingenious models. Although they have shown to be effective, extra information usually increases the cost of data collection, and complex models may result in difficulty in deployment. Innovatively, we avoid the use of extra information or alterations to the model; instead, we fill the transformation space between the observed data and the underlying preferences with randomness. Specifically, we propose a novel model-agnostic and highly generic framework for sequential recommendation called sample enrichment via temporary operations on subsequences (SETO), which temporarily and separately enriches the transformation space via sequence enhancement operations with rationality constraints in training. The transformation space not only exists in the process from input samples to preferences but also in preferences to target samples. We highlight our SETO's effectiveness and versatility over multiple representative and state-of-the-art sequential recommendation models (including six single-domain sequential models and two cross-domain sequential models) across multiple real-world datasets (including three single-domain datasets, three cross-domain datasets and a large-scale industry dataset).

cs.IR↗

arXiv2024

Technique Report of CVPR 2024 PBDL Challenges

The intersection of physics-based vision and deep learning presents an exciting frontier for advancing computer vision technologies. By leveraging the principles of physics to inform and enhance deep learning models, we can develop more robust and accurate vision systems. Physics-based vision aims to invert the processes to recover scene properties such as shape, reflectance, light distribution, and medium properties from images. In recent years, deep learning has shown promising improvements for various vision tasks, and when combined with physics-based vision, these approaches can enhance the robustness and accuracy of vision systems. This technical report summarizes the outcomes of the Physics-Based Vision Meets Deep Learning (PBDL) 2024 challenge, held in CVPR 2024 workshop. The challenge consisted of eight tracks, focusing on Low-Light Enhancement and Detection as well as High Dynamic Range (HDR) Imaging. This report details the objectives, methodologies, and results of each track, highlighting the top-performing solutions and their innovative approaches.

cs.CV↗

arXiv2024

Delocalization of skin steady states

The skin effect, characterized by the tendency of particles to accumulate at the boundaries, has been extensively studied in non-Hermitian systems. In this work, we propose an intuitive Lindbladian composed of two chains with reversed skin localization. The skin steady state is gradually delocalized as the interchain coupling increases. In the single-body scenario, it corresponds to a shift in the scaling of the Liouvillian gap $Δ$ from $Δ\propto N^0$ to $Δ\propto N^{-2}$. Notably, exact diagonalization results reveal a system-size sensitivity of the single-particle Liouvillian spectrum, inherited from the non-Hermitian effective Hamiltonian's system-size sensitivity. We predict that even an arbitrarily small coupling will induce dramatic changes in the Liouvillian spectrum and steady state in the thermodynamic limit, a phenomenon we term the critical Liouvillian skin effect. Additionally, in the many-body scenario, by employing the stochastic Schrödinger equation to unravel the Lindblad master equation, it is revealed that the scaling behavior of steady-state entanglement changes from the area law to the logarithmic law. This work demonstrates the delocalization of both single-body and many-body skin steady states, introducing a novel mechanism for inducing entanglement transitions beyond the quantum Zeno effect.

quant-ph↗

arXiv2024

Exact complex mobility edges and flagellate spectra for non-Hermitian quasicrystals with exponential hoppings

We propose a class of general non-Hermitian quasiperiodic lattice models with exponential hoppings and analytically determine the genuine complex mobility edges by solving its dual counterpart exactly utilizing Avila's global theory. Our analytical formula unveils that the complex mobility edges usually form a loop structure in the complex energy plane. By shifting the eigenenergy a constant $t$, the complex mobility edges of the family of models with different hopping parameter $t$ can be described by a unified formula, formally independent of $t$. By scanning the hopping parameter, we demonstrate the existence of a type of intriguing flagellate-like spectra in complex energy plane, in which the localized states and extended states are well separated by the complex mobility edges. Our result provides a firm ground for understanding the complex mobility edges in non-Hermitian quasiperiodic lattices.

cond-mat.dis-nn↗

arXiv2024

A Survey on Cross-Domain Sequential Recommendation

Cross-domain sequential recommendation (CDSR) shifts the modeling of user preferences from flat to stereoscopic by integrating and learning interaction information from multiple domains at different granularities (ranging from inter-sequence to intra-sequence and from single-domain to cross-domain). In this survey, we first define the CDSR problem using a four-dimensional tensor and then analyze its multi-type input representations under multidirectional dimensionality reductions. Following that, we provide a systematic overview from both macro and micro views. From a macro view, we abstract the multi-level fusion structures of various models across domains and discuss their bridges for fusion. From a micro view, focusing on the existing models, we first discuss the basic technologies and then explain the auxiliary learning technologies. Finally, we exhibit the available public datasets and the representative experimental results as well as provide some insights into future directions for research in CDSR.

cs.IR↗

arXiv2024

MIPI 2024 Challenge on Demosaic for HybridEVS Camera: Methods and Results

The increasing demand for computational photography and imaging on mobile platforms has led to the widespread development and integration of advanced image sensors with novel algorithms in camera systems. However, the scarcity of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). Building on the achievements of the previous MIPI Workshops held at ECCV 2022 and CVPR 2023, we introduce our third MIPI challenge including three tracks focusing on novel image sensors and imaging algorithms. In this paper, we summarize and review the Nighttime Flare Removal track on MIPI 2024. In total, 170 participants were successfully registered, and 14 teams submitted results in the final testing phase. The developed solutions in this challenge achieved state-of-the-art performance on Nighttime Flare Removal. More details of this challenge and the link to the dataset can be found at https://mipi-challenge.org/MIPI2024/.

eess.IV↗