Search

Class-incremental learning (CIL) aims to acquire new classes while conserving historical knowledge incrementally. Despite existing pre-trained model (PTM) based methods performing excellently in CIL, it is better to fine-tune them on downstream incremental tasks with massive patterns unknown to PTMs. However, using task streams for fine-tuning could lead to catastrophic forgetting that will erase the knowledge in PTMs. This paper proposes the Dual Prototype network for Task-wise Adaption (DPTA) of PTM-based CIL. For each incremental learning task, a task-wise adapter module is built to fine-tune the PTM, where the center-adapt loss forces the representation to be more centrally clustered and class separable. The dual prototype network improves the prediction process by enabling test-time adapter selection, where the raw prototypes deduce several possible task indexes of test samples to select suitable adapter modules for PTM, and the augmented prototypes that could separate highly correlated classes are utilized to determine the final result. Experiments on several benchmark datasets demonstrate the state-of-the-art performance of DPTA. The code will be open-sourced after the paper is published., Comment: 8 pages,6 figures,2 tables

Deep learning electronic structures from ab initio calculations holds great potential to revolutionize computational materials studies. While existing methods proved success in deep-learning density functional theory (DFT) Hamiltonian matrices, they are limited to DFT programs using localized atomic-like bases and heavily depend on the form of the bases. Here, we propose the DeepH-r method for deep-learning DFT Hamiltonians in real space, facilitating the prediction of DFT Hamiltonian in a basis-independent manner. An equivariant neural network architecture for modeling the real-space DFT potential is developed, targeting a more fundamental quantity in DFT. The real-space potential exhibits simplified principles of equivariance and enhanced nearsightedness, further boosting the performance of deep learning. When applied to evaluate the Hamiltonian matrix, this method significantly improved in accuracy, as exemplified in multiple case studies. Given the abundance of data in the real-space potential, this work may pave a novel pathway for establishing a ``large materials model" with increased accuracy.

Realizing large materials models has emerged as a critical endeavor for materials research in the new era of artificial intelligence, but how to achieve this fantastic and challenging objective remains elusive. Here, we propose a feasible pathway to address this paramount pursuit by developing universal materials models of deep-learning density functional theory Hamiltonian (DeepH), enabling computational modeling of the complicated structure-property relationship of materials in general. By constructing a large materials database and substantially improving the DeepH method, we obtain a universal materials model of DeepH capable of handling diverse elemental compositions and material structures, achieving remarkable accuracy in predicting material properties. We further showcase a promising application of fine-tuning universal materials models for enhancing specific materials models. This work not only demonstrates the concept of DeepH's universal materials model but also lays the groundwork for developing large materials models, opening up significant opportunities for advancing artificial intelligence-driven materials discovery.

Recent experimental discovery of fractional Chern insulator in moir\'e Chern band in twisted transition metal dichalocogenide homobilayers has sparked intensive interest in exploring the ways of engineering band topology and correlated states in moir\'e systems. In this letter, we demonstrate that, with an additional exchange interaction induced by proximity effect, the topology and bandwidth of the moir\'e minibands of twisted $\mathrm{MoTe_2}$ homobilayers can be easily tuned. Fractional Chern insulators at -2/3 filling are found to appear at enlarged twist angles over a large range of twist angles with enhanced many-body gaps. We further discover a topological phase transition between the fractional Chern insulator, quantum anomalous Hall crystal, and charge density wave. Our results shed light on the interplay between topology and correlation physics., Comment: 7 pages, 4 figures

Antiferromagnetic (AFM) skyrmions, which are resistant to both the skyrmion Hall effect and external magnetic perturbations, are expected to be promising candidates for next-generation spintronics devices. Despite being observed in bulk materials and synthetic AFM layered systems, the existence of intrinsic AFM skyrmions within single magnetic layers, which offer potential advantages for spintronic device fabrication, has remained elusive. In this work, taking monolayer CrSi(Te,Se)$_{3}$ as a representative system, we demonstrate the emergence of intrinsic AFM skyrmions in two-dimensional Janus magnets. It is found that under moderate compressive strain, the interplay between considerable Dyzaloshinskii-Moriya interaction and the strain-induced AFM Heisenberg exchange interaction in monolayer CrSi(Te,Se)$_{3}$ would give rise to the emergence of intrinsic AFM skyrmions assembled from AFM spin spirals. Moreover, the application of an external magnetic field could trigger the emergence of AFM merons as well as a canted AFM state. Our findings propose a feasible approach for achieving intrinsic AFM skyrmions in realistic systems, which paves the way for developments in AFM topological spintronics devices., Comment: 10 pages, 6 figures

Neural network force fields have significantly advanced ab initio atomistic simulations across diverse fields. However, their application in the realm of magnetic materials is still in its early stage due to challenges posed by the subtle magnetic energy landscape and the difficulty of obtaining training data. Here we introduce a data-efficient neural network architecture to represent density functional theory total energy, atomic forces, and magnetic forces as functions of atomic and magnetic structures. Our approach incorporates the principle of equivariance under the three-dimensional Euclidean group into the neural network model. Through systematic experiments on various systems, including monolayer magnets, curved nanotube magnets, and moir\'e-twisted bilayer magnets of $\text{CrI}_{3}$, we showcase the method's high efficiency and accuracy, as well as exceptional generalization ability. The work creates opportunities for exploring magnetic phenomena in large-scale materials systems., Comment: 10 pages, 4 figures

This paper addresses spoken language understanding (SLU) on microcontroller-like embedded devices, integrating on-device execution with cloud offloading in a novel fashion. We leverage temporal locality in the speech inputs to a device and reuse recent SLU inferences accordingly. Our idea is simple: let the device match incoming inputs against cached results, and only offload inputs not matched to any cached ones to the cloud for full inference. Realization of this idea, however, is non-trivial: the device needs to compare acoustic features in a robust yet low-cost way. To this end, we present SpeechCache (or SC), a speech cache for tiny devices. It matches speech inputs at two levels of representations: first by sequences of clustered raw sound units, then as sequences of phonemes. Working in tandem, the two representations offer complementary tradeoffs between cost and efficiency. To boost accuracy even further, our cache learns to personalize: with the mismatched and then offloaded inputs, it continuously finetunes the device's feature extractors with the assistance of the cloud. We implement SC on an off-the-shelf STM32 microcontroller. The complete implementation has a small memory footprint of 2MB. Evaluated on challenging speech benchmarks, our system resolves 45%-90% of inputs on device, reducing the average latency by up to 80% compared to offloading to popular cloud speech recognition services. The benefit brought by our proposed SC is notable even in adversarial settings - noisy environments, cold cache, or one device shared by a number of users., Comment: accepted at MobiSys'24

Chiral magnetic states in two-dimensional (2D) layered noncentrosymmetric magnets, which are promising advanced spintronic materials, are usually attributed to Dzyaloshinskii-Moriya interactions (DMI). However, the role of underlying higher-order spin couplings in determining the properties of chiral spin textures has much less reported. In this work, taking the lithium-decorated monolayer CrTe$_{2}$ (monolayer LiCrTe$_{2}$) as an example, we develop a first-principles-based comprehensive spin model constructed by using the symmetry-adapted cluster expansion method. Based on this spin model, we identify the ground state of monolayer LiCrTe$_{2}$ as a chiral spin spiral state, which can further assemble macroscopic chiral labyrinth domains (LD) under zero-field conditions as well as evolve into skyrmions under a finite magnetic field. Moreover, higher-order biquadratic and three-site interactions are identified to be responsible for modulating both the size and the field stability of the spin spiral state. Our study sheds light on complex magnetic couplings in 2D magnets., Comment: 10 pages, 5 figures

Atomically thin van der Waals magnetic materials have not only provided a fertile playground to explore basic physics in the two-dimensional (2D) limit but also created vast opportunities for novel ultrafast functional devices. Here we systematically investigate ultrafast magnetization dynamics and spin wave dynamics in few-layer topological antiferromagnetic MnBi2Te4 crystals as a function of layer number, temperature, and magnetic field. We find laser-induced (de)magnetization processes can be used to accurately track the distinct magnetic states in different magnetic field regimes, including showing clear odd-even layer number effects. In addition, strongly field-dependent antiferromagnetic magnon modes with tens of gigahertz frequencies are optically generated and directly observed in the time domain. Remarkably, we find that magnetization and magnon dynamics can be observed in not only the time-resolved magneto-optical Kerr effect but also the time resolved reflectivity, indicating strong correlation between the magnetic state and electronic structure. These measurements present the first comprehensive overview of ultrafast spin dynamics in this novel 2D antiferromagnet, paving the way for potential applications in 2D antiferromagnetic spintronics and magnonics as well as further studies of ultrafast control of both magnetization and topological quantum states.

The quantum anomalous Hall effect (QAHE) has unique advantages in topotronic applications, but it is still challenging to realize the QAHE with tunable magnetic and topological properties for building functional devices. Through systematic first-principles calculations, we predict that the in-plane magnetization induced QAHE with Chern numbers C = $\pm$1 and the out-of-plane magnetization induced QAHE with high Chern numbers C = $\pm$3 can be realized in a single material candidate, which is composed of van der Waals (vdW) coupled Bi and MnBi$_2$Te$_4$ monolayers. The switching between different phases of QAHE can be controllable by multiple ways, such as applying strain or (weak) magnetic field or twisting the vdW materials. The prediction of an experimentally available material system hosting robust, highly tunable QAHE will stimulate great research interest in the field. Our work opens a new avenue for the realization of tunable QAHE and provides a practical material platform for the development of topological electronics., Comment: 14 pages, 4 figures

Finding guiding principles to optimize properties of quantum anomalous Hall (QAH) insulators is of pivotal importance to fundamental science and applications. Here, we build a first-principles QAH material database of chirality and band gap, explore microscopic mechanisms determining the QAH material properties, and obtain a general physical picture that can comprehensively understand the QAH data. Our results reveal that the usually neglected Coulomb exchange is unexpectedly strong in a large class of QAH materials, which is the key to resolve experimental puzzles. Moreover, we identify simple indicators for property evaluation and suggest material design strategies to control QAH chirality and gap by tuning cooperative or competing contributions via magnetic co-doping, heterostructuring, spin-orbit proximity, etc. The work is valuable to future research of magnetic topological physics and materials.

Recently, the intrinsic magnetic topological insulator MnBi2Te4 has attracted enormous research interest due to the great success in realizing exotic topological quantum states, such as the quantum anomalous Hall effect (QAHE), axion insulator state, high-Chern-number and high-temperature Chern insulator states. One key issue in this field is to effectively manipulate these states and control topological phase transitions. Here, by systematic angle-dependent transport measurements, we reveal a magnetization-tuned topological quantum phase transition from Chern insulator to magnetic insulator with gapped Dirac surface states in MnBi2Te4 devices. Specifically, as the magnetic field is tilted away from the out-of-plane direction by around 40-60 degrees, the Hall resistance deviates from the quantization value and a colossal, anisotropic magnetoresistance is detected. The theoretical analyses based on modified Landauer-Buttiker formalism show that the field-tilt-driven switching from ferromagnetic state to canted antiferromagnetic state induces a topological quantum phase transition from Chern insulator to magnetic insulator with gapped Dirac surface states in MnBi2Te4 devices. Our work provides an efficient means for modulating topological quantum states and topological quantum phase transitions.