基于ALBERT、Lattice IndyLSTM和Attention融合的用户意图分类

旨在对话式智能系统中的用户意图识别任务,论文提出了一种新型的用户意图分类模型。该模型结合了ALBERT预训练模型、格形式的独立循环长短期记忆网络(Lattice IndyLSTM)以及词注意力融合机制。通过构造一个由字嵌入和词嵌入组成的Lattice,输入至IndyLSTM网络实现对用户输入语句编码,能够处理传统意图分类任务中无法同时利用字词信息、RNN中存在梯度爆炸或者消失,LSTM模型过拟合等问题。此外,利用能够提高对用户输入语句中领域特定词汇的编码贡献度的词注意力机制,大大提升了意图识别准确性。通过实验得出,论文提出的用户意图分类模型能够有效提升精确率、召回率和F1值等指标。

Exciting lattice oxygen of nickel–iron bi-metal alkoxide for efficient electrochemical oxygen evolution reaction

High efficiency,cost-effective and durable electrocatalysts are of pivotal importance in energy conversion and storage systems.The electro-oxidation of water to oxygen plays a crucial role in such energy conversion technologies.Herein,we report a robust method for the synthesis of a bimetallic alkoxide for efficient oxygen evolution reaction(OER)for alkaline electrolysis,which yields current density of 10 mA cm^(-2)at an overpotential of 215 mV in 0.1 M KOH electrolyte.The catalyst demonstrates an excellent durability for more than 540 h operation with negligible degradation in activity.Raman spectra revealed that the catalyst underwent structure reconstruction during OER,evolving into oxyhydroxide,which was the active site proceeding OER in alkaline electrolyte.In-situ synchrotron X-ray absorption experiment combined with density functional theory calculation suggests a lattice oxygen involved electrocatalytic reaction mechanism for the in-situ generated nickel–iron bimetal-oxyhydroxide catalyst.This mechanism together with the synergy between nickel and iron are responsible for the enhanced catalytic activity and durability.These findings provide promising strategies for the rational design of nonnoble metal OER catalysts.

An inverse analysis of fluid flow through granular media using differentiable lattice Boltzmann method

This study presents a method for the inverse analysis of fluid flow problems.The focus is put on accurately determining boundary conditions and characterizing the physical properties of granular media,such as permeability,and fluid components,like viscosity.The primary aim is to deduce either constant pressure head or pressure profiles,given the known velocity field at a steady-state flow through a conduit containing obstacles,including walls,spheres,and grains.The lattice Boltzmann method(LBM)combined with automatic differentiation(AD)(AD-LBM)is employed,with the help of the GPU-capable Taichi programming language.A lightweight tape is used to generate gradients for the entire LBM simulation,enabling end-to-end backpropagation.Our AD-LBM approach accurately estimates the boundary conditions for complex flow paths in porous media,leading to observed steady-state velocity fields and deriving macro-scale permeability and fluid viscosity.The method demonstrates significant advantages in terms of prediction accuracy and computational efficiency,making it a powerful tool for solving inverse fluid flow problems in various applications.

Volumetric lattice Boltzmann method for pore-scale mass diffusionadvection process in geopolymer porous structures

Porous materials present significant advantages for absorbing radioactive isotopes in nuclear waste streams.To improve absorption efficiency in nuclear waste treatment,a thorough understanding of the diffusion-advection process within porous structures is essential for material design.In this study,we present advancements in the volumetric lattice Boltzmann method(VLBM)for modeling and simulating pore-scale diffusion-advection of radioactive isotopes within geopolymer porous structures.These structures are created using the phase field method(PFM)to precisely control pore architectures.In our VLBM approach,we introduce a concentration field of an isotope seamlessly coupled with the velocity field and solve it by the time evolution of its particle population function.To address the computational intensity inherent in the coupled lattice Boltzmann equations for velocity and concentration fields,we implement graphics processing unit(GPU)parallelization.Validation of the developed model involves examining the flow and diffusion fields in porous structures.Remarkably,good agreement is observed for both the velocity field from VLBM and multiphysics object-oriented simulation environment(MOOSE),and the concentration field from VLBM and the finite difference method(FDM).Furthermore,we investigate the effects of background flow,species diffusivity,and porosity on the diffusion-advection behavior by varying the background flow velocity,diffusion coefficient,and pore volume fraction,respectively.Notably,all three parameters exert an influence on the diffusion-advection process.Increased background flow and diffusivity markedly accelerate the process due to increased advection intensity and enhanced diffusion capability,respectively.Conversely,increasing the porosity has a less significant effect,causing a slight slowdown of the diffusion-advection process due to the expanded pore volume.This comprehensive parametric study provides valuable insights into the kinetics of isotope uptake in porous structures,facilitating the development of porous materials for nuclear waste treatment applications.

Observation of flat-band localized state in a one-dimensional diamond momentum lattice of ultracold atoms

We investigated the one-dimensional diamond ladder in the momentum lattice platform. By inducing multiple twoand four-photon Bragg scatterings among specific momentum states, we achieved a flat band system based on the diamond model, precisely controlling the coupling strength and phase between individual lattice sites. Utilizing two lattice sites couplings, we generated a compact localized state associated with the flat band, which remained localized throughout the entire time evolution. We successfully realized the continuous shift of flat bands by adjusting the corresponding nearest neighbor hopping strength, enabling us to observe the complete localization process. This opens avenues for further exploration of more complex properties within flat-band systems, including investigating the robustness of flat-band localized states in disordered flat-band systems and exploring many-body localization in interacting flat-band systems.

A novel triple periodic minimal surface-like plate lattice and its data-driven optimization method for superior mechanical properties

Lattice structures can be designed to achieve unique mechanical properties and have attracted increasing attention for applications in high-end industrial equipment,along with the advances in additive manufacturing(AM)technologies.In this work,a novel design of plate lattice structures described by a parametric model is proposed to enrich the design space of plate lattice structures with high connectivity suitable for AM processes.The parametric model takes the basic unit of the triple periodic minimal surface(TPMS)lattice as a skeleton and adopts a set of generation parameters to determine the plate lattice structure with different topologies,which takes the advantages of both plate lattices for superior specific mechanical properties and TPMS lattices for high connectivity,and therefore is referred to as a TPMS-like plate lattice(TLPL).Furthermore,a data-driven shape optimization method is proposed to optimize the TLPL structure for maximum mechanical properties with or without the isotropic constraints.In this method,the genetic algorithm for the optimization is utilized for global search capability,and an artificial neural network(ANN)model for individual fitness estimation is integrated for high efficiency.A set of optimized TLPLs at different relative densities are experimentally validated by the selective laser melting(SLM)fabricated samples.It is confirmed that the optimized TLPLs could achieve elastic isotropy and have superior stiffness over other isotropic lattice structures.

Certificateless threshold signature from lattice

Threshold signature has been widely used in electronic wills,electronic elections,cloud computing,secure multiparty computation and other fields.Until now,certificateless threshold signature schemes are all based on traditional mathematic theory,so they cannot resist quantum computing attacks.In view of this,we combine the advantages of lattice-based cryptosystem and certificateless cryptosystem to construct a certificateless threshold signature from lattice(LCLTS)that is efficient and resistant to quantum algorithm attacks.LCLTS has the threshold characteristics and can resist the quantum computing attacks,and the analysis shows that it is unforgeable against the adaptive Chosen-Message Attacks(UF-CMA)with the difficulty of Inhomogeneous Small Integer Solution(ISIS)problem.In addition,LCLTS solves the problems of the certificate management through key escrow.

Effects of Rattling Behavior of K and Cd Atoms along Different Directions in Anisotropic KCdAs on Lattice Thermal Transport and Thermoelectric Properties

We employ advanced first principles methodology,merging self-consistent phonon theory and the Boltzmann transport equation,to comprehensively explore the thermal transport and thermoelectric properties of KCdAs.Notably,the study accounts for the impact of quartic anharmonicity on phonon group velocities in the pursuit of lattice thermal conductivity and investigates 3ph and 4ph scattering processes on phonon lifetimes.Through various methodologies,including examining atomic vibrational modes and analyzing 3ph and 4ph scattering processes,the article unveils microphysical mechanisms contributing to the lowκL within KCdAs.Key features include significant anisotropy in Cd atoms,pronounced anharmonicity in K atoms,and relative vibrations in non-equivalent As atomic layers.Cd atoms,situated between As layers,exhibit rattling modes and strong lattice anharmonicity,contributing to the observed lowκL.Remarkably flat bands near the valence band maximum translate into high PF,aligning with ultralowκL for exceptional thermoelectric performance.Under optimal temperature and carrier concentration doping,outstanding ZT values are achieved:4.25(a(b)-axis,p-type,3×10^(19)cm^(−3),500 K),0.90(c-axis,p-type,5×10^(20)cm^(−3),700 K),1.61(a(b)-axis,n-type,2×10^(18)cm^(−3),700 K),and 3.06(c-axis,n-type,9×10^(17)cm^(−3),700 K).

On Discrete Hopf Fibrations, Grand Unification Groups, the Barnes-Wall, Leech Lattices, and Quasicrystals

A discrete Hopf fibration of S15 over S8 with S7 (unit octonions) as fibers leads to a 16D Polytope P16 with 4320 vertices obtained from the convex hull of the 16D Barnes-Wall lattice Λ16. It is argued (conjectured) how a subsequent 2-1 mapping (projection) of P16 onto a 8D-hyperplane might furnish the 2160 vertices of the uniform 241 polytope in 8-dimensions, and such that one can capture the chain sequence of polytopes 241,231,221,211in D=8,7,6,5dimensions, leading, respectively, to the sequence of Coxeter groups E8,E7,E6,SO(10)which are putative GUT group candidates. An embedding of the E8⊕E8and E8⊕E8⊕E8lattice into the Barnes-Wall Λ16 and Leech Λ24 lattices, respectively, is explicitly shown. From the 16D lattice E8⊕E8one can generate two separate families of Elser-Sloane 4D quasicrystals (QC’s) with H4 (icosahedral) symmetry via the “cut-and-project” method from 8D to 4D in each separate E8 lattice. Therefore, one obtains in this fashion the Cartesian product of two Elser-Sloane QC’s Q×Qspanning an 8D space. Similarly, from the 24D lattice E8⊕E8⊕E8one can generate the Cartesian product of three Elser-Sloane 4D quasicrystals (QC’s) Q×Q×Qwith H4 symmetry and spanning a 12D space.

Band structures of strained kagome lattices

Materials with kagome lattices have attracted significant research attention due to their nontrivial features in energy bands.We theoretically investigate the evolution of electronic band structures of kagome lattices in response to uniaxial strain using both a tight-binding model and an antidot model based on a periodic muffin-tin potential.It is found that the Dirac points move with applied strain.Furthermore,the flat band of unstrained kagome lattices is found to develop into a highly anisotropic shape under a stretching strain along y direction,forming a partially flat band with a region dispersionless along ky direction while dispersive along kx direction.Our results shed light on the possibility of engineering the electronic band structures of kagome materials by mechanical strain.

On the spreading behavior of a droplet on a circular cylinder using the lattice Boltzmann method

The study of a droplet spreading on a circular cylinder under gravity was carried out using the pseudo-potential lattice Boltzmann high-density ratios multiphase model with a non-ideal Peng–Robinson equation of state. The calculation results indicate that the motion of the droplet on the cylinder can be divided into three stages: spreading, sliding, and aggregating.The contact length and contact time of a droplet on a cylindrical surface can be affected by factors such as the wettability gradient of the cylindrical wall, the Bond number, and droplet size. Furthermore, phase diagrams showing the relationship between Bond number, cylinder wall wettability gradient, and contact time as well as maximum contact length for three different droplet sizes are given. A theoretical foundation for additional research into the heat and mass transfer process between the droplet and the cylinder can be established by comprehending the variable rules of maximum contact length and contact time.

Speed limit effect during lane change in a two-lane lattice model under V2X environment

Speed limit measures are ubiquitous due to the complexity of the road environment,which can be supplied with the help of vehicle to everything(V2X)communication technology.Therefore,the influence of speed limit on traffic system will be investigated to construct a two-lane lattice model accounting for the speed limit effect during the lane change process under V2X environment.Accordingly,the stability condition and the mKdV equation are closely associated with the speed limit effect through theory analysis.Moreover,the evolution of density and hysteresis loop is simulated to demonstrate the positive role of the speed limit effect on traffic stability in the cases of strong reaction intensity and high limited speed.

{1012}twin–twin intersection-induced lattice rotation and dynamic recrystallization in Mg–6Al–3Sn–2Zn alloy

This study investigated the formation mechanism of new grains due to twin–twin intersections in a coarse-grained Mg–6Al–3Sn–2Zn alloy during different strain rates of an isothermal compression.The results of electron backscattered diffraction investigations showed that the activated twins were primarily{1012}tension twins,and 60°<1010>boundaries formed due to twin–twin intersections under different strain rates.Isolated twin variants with 60°<1010>boundaries transformed into new grains through lattice rotations at a low strain rate(0.01 s^(−1)).At a high strain rate(10 s^(−1)),the regions surrounded by subgrain boundaries through high-density dislocation arrangement and the 60°<1010>boundaries transformed into new grains via dynamic recrystallization.

基于LATTICE的文言文自动断句研究

随着社会主义文化强国的建设,利用自然语言处理技术实现对古汉语典籍的处理与挖掘越发受到关注。针对文言文自动化断句任务,论文提出一种基于BERT-LATTICE-LAN网络的深度学习模型,通过集成预训练模型和融入词向量的长短时记忆网络,实现对文言文语句的自动化断句处理。相比于传统的BiLSTM-CRF网络,其准确率和F值分别提高了8.03%和7.88%。

Operando measurement of lattice deformation profiles of synchrotron radiation monochromator

This study presents a new method for characterizing the thermal lattice deformation of a monochromator with high precision under service conditions and first reports the operando measurements of nanoscale thermal lattice deformation on a double-crystal monochromator at different incident powers.The nanoscale thermal lattice deformation of the monochromator first crystal was obtained by analyzing the intensity of the distorted DuMond diagrams.DuMond diagrams of the 333 diffraction index,sensitive to lattice deformation,were obtained directly using a 2D detector and an analyzer crystal orthogonal to the monochromator.With increasing incident power and power density,the maximum height of the lattice deformation increased from 3.2 to 18.5 nm,and the deformation coefficient of the maximum height increased from 1.1 to 3.2 nm/W.The maximum relative standard deviation was 4.2%,and the maximum standard deviation was 0.1 nm.Based on the measured thermal deformations,the flux saturation phenomenon and critical point for the linear operation of the monochromator were predicted with increasing incident power.This study provides a simple solution to the problem of the lower precision of synchrotron radiation monochromator characterizations compared to simulations.

Synergism of preintercalated manganese ions and lattice water in vanadium oxide cathodes for high-capacity and long-life Zn-ion batteries

Aqueous Zn-ion batteries(AZIBs)are recognized as a promising energy storage system with intrinsic safety and low cost,but its applications still rely on the design of high-capacity and stable-cycling cathode materials.In this work,we present an intercalation mechanism-based cathode materials for AZIB,i.e.the vanadium oxide with pre-intercalated manganese ions and lattice water(noted as MVOH).The synergistic effect between Mn^(2+)and lattice H_(2)O not only expands the interlayer spacing,but also significantly enhances the structural stability.Systematic in-situ and ex-situ characterizations clarify the Zn^(2+)/H^(+)co–(de)intercalation mechanism of MVOH in aqueous electrolyte.The demonstrated remarkable structure stability,excellent kinetic behaviors and ion-storage mechanism together enable the MVOH to demonstrate satisfactory specific capacity of 450 mA h g^(−1)at 0.2 A g^(−1),excellent rate performance of 288.8 mA h g^(−1)at 10 A g^(−1)and long cycle life over 20,000 cycles at 5 A g^(−1).This work provides a practical cathode material,and contributes to the understanding of the ion-intercalation mechanism and structural evolution of the vanadium-based cathode for AZIBs.

Valley switch effect in anα-T3 lattice-based superconducting interferometer

Dirac electrons possess a valley degree of freedom,which is currently under investigation as a potential information carrier.We propose an approach to generate and manipulate the valley-switching current(VSC)through Andreev reflection using an interferometer-based superconductor hybrid junction.The interferometer comprises a ring-shaped structure formed by topological kink states in the a-T3 lattice via carefully designed electrostatic potentials.Our results demonstrate the feasibility of achieving a fully polarized VSC in this device without contamination from cotunneling electrons sharing the same valley as the incident electron.Furthermore,we show that control over the fully polarized VSC can be achieved by applying a nonlocal gate voltage or modifying the global parameter a.The former alters the dynamic phase of electrons while the latter provides an a-dependent Berry phase,both directly influencing quantum interference and thereby affecting performance in terms of generating and manipulating VSC,crucial for advancements in valleytronics.

Asymmetric configuration activating lattice oxygen via weakening d-p orbital hybridization for efficient C/N separation in urea overall electrolysis

Urea oxidation reaction(UOR)is proposed as an exemplary half-reaction in renewable energy applications because of its low thermodynamical potential.However,challenges persist due to sluggish reaction kinetics and complex by-products separation.To this end,we introduce the lattice oxygen oxidation mechanism(LOM),propelling a novel UOR route using a modified CoFe layered double hydroxide(LDH)catalyst termed CFRO-7.Theoretical calculations and in-situ characterizations highlight the activated lattice oxygen(O_(L))within CFRO-7 as pivotal sites for UOR,optimizing the reaction pathway and accelerating the kinetics.For the urea overall electrolysis application,the LOM route only requires a low voltage of 1.54 V to offer a high current of 100 mA cm^(-2) for long-term utilization(>48 h).Importantly,the by-product NCO^(-)−is significantly suppressed,while the CO_(2)2/N_(2) separation is efficiently achieved.This work proposed a pioneering paradigm,invoking the LOM pathway in urea electrolysis to expedite reaction dynamics and enhance product selectivity.

A Yb optical clock with a lattice power enhancement cavity

We construct a power enhancement cavity to form an optical lattice in an ytterbium optical clock.It is demonstrated that the intra-cavity lattice power can be increased by about 45 times,and the trap depth can be as large as 1400Er when laser light with a power of only 0.6 W incident to the lattice cavity.Such high trap depths are the key to accurate evaluation of the lattice-induced light shift with an uncertainty down to~1×10-18.By probing the ytterbium atoms trapped in the power-enhanced optical lattice,we obtain a 4.3 Hz-linewidth Rabi spectrum,which is then used to feedback to the clock laser for the close loop operation of the optical lattice clock.We evaluate the density shift of the Yb optical lattice clock based on interleaving measurements,which is-0.46(62)mHz.This result is smaller compared to the density shift of our first Yb optical clock without lattice power enhancement cavity mainly due to a larger lattice diameter of 344μm.

Local failure mechanism of sand-blocking fence in latticed dune along desert roads

The latticed dunes in the Tengger Desert are widely distributed,and the sand-blocking fence placed here are highly susceptible to local failure due to complex wind-sand activities,posing a serious threat to the safe operation of the highway.To explore the local failure mechanism of sand-blocking fence in the latticed dune area,the local failure of sand-blocking fence in the latticed dune areas along the Wuhai-Maqin Highway in China was observed.Taking the first main ridge of the latticed dune as the placement location,the structure of the wind-sand flow field of sand-blocking fence placed at top,the bottom and the middle of windward slope was analyzed by Computational Fluid Dynamics(CFD).The results show that when placed at top of the first main ridge,the wind speed near the sand-blocking fence is the highest,up to 15.23 m/s.Therefore,the wind load strength on the sand barrier is correspondingly larger,up to 232.61 N∙m-2.As the strength of material continues to decrease,the nylon net is prone to breakage.The roots of the angle steel posts are susceptible to hollowing by vortex action,which can cause sand-blocking fence to fall over in strong wind conditions.When placed at the bottom of windward slope,wind speed drop near sand-blocking fence is greatest,with the decrease of 12.48-14.32 m/s compared to the original wind speed.This is highly likely to lead to large-scale deposition of sand particles and burial of the sand-blocking fence.When placed in the middle of windward slope,sand-blocking fence is subjected to less wind load strength(168.61N∙m-2)and sand particles are mostly deposited at the bottom of windward slope,with only a small amount of sand accumulating at the root of sand-blocking fence.Based on field observations and numerical modelling results,when the sand-blocking fence is placed in latticed dune area,it should be placed in the middle of the windward slope of the first main ridge as a matter of priority.Besides the sand-blocking fence should be placed at the top of the first main ridge,and sand fixing measures should be added.