Dirac semimetals lack a simple bulk-boundary correspondence. Recently, Dirac materials with four-fold rotation symmetry have been shown to exhibit a higher order bulk-hinge correspondence: they display higher order Fermi arcs, which are localized on
hinges where two surfaces meet and connect the projections of the bulk Dirac points. In this paper, we classify higher order Fermi arcs for Dirac semimetals protected by a rotation symmetry and the product of time-reversal and inversion. Such Dirac points can be either linear in all directions or linear along the rotation axis and quadratic in other directions. By computing the filling anomaly for momentum-space planes on either side of the Dirac point, we find that all linear Dirac points exhibit higher order Fermi arcs terminating at the projection of the Dirac point, while the Dirac points that are quadratic in two directions lack such higher order Fermi arcs. When higher order Fermi arcs do exist, they obey either a $mathbb{Z}_2$ (four-fold rotation axis) or $mathbb{Z}_3$ (three- or six-fold rotation axis) group structure. Finally, we build two models with six-fold symmetry to illustrate the cases with and without higher order Fermi arcs. We predict higher order Fermi arcs in Na$_3$Bi.
Origin-Destination (OD) flow, as an abstract representation of the object`s movement or interaction, has been used to reveal the urban mobility and human-land interaction pattern. As an important spatial analysis approach, the clustering methods of p
oint events have been extended to OD flows to identify the dominant trends and spatial structures of urban mobility. However, the existing methods for OD flow cluster-detecting are limited both in specific spatial scale and the uncertain result due to different parameters setting, which is difficult for complicated OD flows clustering under spatial heterogeneity. To address these limitations, in this paper, we proposed a novel OD flows cluster-detecting method based on the OPTICS algorithm which can identify OD flow clusters with various aggregation scales. The method can adaptively determine parameter value from the dataset without prior knowledge and artificial intervention. Experiments indicated that our method outperformed three state-of-the-art methods with more accurate and complete of clusters and less noise. As a case study, our method is applied to identify the potential routes for public transport service settings by detecting OD flow clusters within urban travel data.
The superconducting order parameter of the first heavy-fermion superconductor CeCu2Si2 is currently under debate. A key ingredient to understand its superconductivity and physical properties is the quasiparticle dispersion and Fermi surface, which re
mains elusive experimentally. Here we present measurements from angle-resolved photoemission spectroscopy. Our results emphasize the key role played by the Ce 4f electrons for the low-temperature Fermi surface, highlighting a band-dependent conduction-f electron hybridization. In particular, we find a very heavy quasi-two-dimensional electron band near the bulk X point and moderately heavy three-dimensional hole pockets near the Z point. Comparison with theoretical calculations reveals the strong local correlation in this compound, calling for further theoretical studies. Our results provide the electronic basis to understand the heavy fermion behavior and superconductivity; implications for the enigmatic superconductivity of this compound are also discussed.
Semi-supervised node classification on graphs is an important research problem, with many real-world applications in information retrieval such as content classification on a social network and query intent classification on an e-commerce query graph
. While traditional approaches are largely transductive, recent graph neural networks (GNNs) integrate node features with network structures, thus enabling inductive node classification models that can be applied to new nodes or even new graphs in the same feature space. However, inter-graph differences still exist across graphs within the same domain. Thus, training just one global model (e.g., a state-of-the-art GNN) to handle all new graphs, whilst ignoring the inter-graph differences, can lead to suboptimal performance. In this paper, we study the problem of inductive node classification across graphs. Unlike existing one-model-fits-all approaches, we propose a novel meta-inductive framework called MI-GNN to customize the inductive model to each graph under a meta-learning paradigm. That is, MI-GNN does not directly learn an inductive model; it learns the general knowledge of how to train a model for semi-supervised node classification on new graphs. To cope with the differences across graphs, MI-GNN employs a dual adaptation mechanism at both the graph and task levels. More specifically, we learn a graph prior to adapt for the graph-level differences, and a task prior to adapt for the task-level differences conditioned on a graph. Extensive experiments on five real-world graph collections demonstrate the effectiveness of our proposed model.
In this paper, we study the semilinear wave equations with the inverse-square potential. By transferring the original equation to a fractional dimensional wave equation and analyzing the properties of its fundamental solution, we establish a long-tim
e existence result, for sufficiently small, spherically symmetric initial data. Together with the previously known blow-up result, we determine the critical exponent which divides the global existence and finite time blow-up. Moreover, the sharp lower bounds of the lifespan are obtained, except for certain borderline case. In addition, our technology allows us to handle an extreme case for the potential, which has hardly been discussed in literature.
Despite exciting progress in pre-training for visual-linguistic (VL) representations, very few aspire to a small VL model. In this paper, we study knowledge distillation (KD) to effectively compress a transformer-based large VL model into a small VL
model. The major challenge arises from the inconsistent regional visual tokens extracted from different detectors of Teacher and Student, resulting in the misalignment of hidden representations and attention distributions. To address the problem, we retrain and adapt the Teacher by using the same region proposals from Students detector while the features are from Teachers own object detector. With aligned network inputs, the adapted Teacher is capable of transferring the knowledge through the intermediate representations. Specifically, we use the mean square error loss to mimic the attention distribution inside the transformer block and present a token-wise noise contrastive loss to align the hidden state by contrasting with negative representations stored in a sample queue. To this end, we show that our proposed distillation significantly improves the performance of small VL models on image captioning and visual question answering tasks. It reaches 120.8 in CIDEr score on COCO captioning, an improvement of 5.1 over its non-distilled counterpart; and an accuracy of 69.8 on VQA 2.0, a 0.8 gain from the baseline. Our extensive experiments and ablations confirm the effectiveness of VL distillation in both pre-training and fine-tuning stages.
The 4f-electron delocalization plays a key role in the low-temperature properties of rare-earth metals and intermetallics, including heavy fermions and mix-valent compounds, and is normally realized by the many-body Kondo coupling between 4f and cond
uction electrons. Due to the large onsite Coulomb repulsion of 4f electrons, the bandwidth-control Mott-type delocalization, commonly observed in d-electron systems, is difficult in 4f-electron systems and remains elusive in spectroscopic experiments. Here we demonstrate that the bandwidth-control orbital-selective delocalization of 4f electrons can be realized in epitaxial Ce films by thermal annealing, which results in a metastable surface phase with a reduced layer spacing. The resulting quasiparticle bands exhibit large dispersion with exclusive 4f character near E_F and extend reasonably far below the Fermi energy, which can be explained from the Mott physics. The experimental quasiparticle dispersion agrees surprisingly well with density-functional theory calculation and also exhibits unusual temperature dependence, which could be a direct consequence of the delicate interplay between the bandwidth-control Mott physics and the coexisting Kondo hybridization. Our work therefore opens up the opportunity to study the interaction between two well-known localization-delocalization mechanisms in correlation physics, i.e., Kondo vs Mott, which can be important for a fundamental understanding of 4f-electron systems.
We report growth, electronic structure and superconductivity of ultrathin epitaxial CoSi2 films on Si(111). At low coverages, preferred islands with 2, 5 and 6 monolayers height develop, which agrees well with the surface energy calculation. We obser
ve clear quantum well states as a result of electronic confinement and their dispersion agrees well with density functional theory calculations, indicating weak correlation effect despite strong contributions from Co 3d electrons. Ex-situ transport measurements show that superconductivity persists down to at least 10 monolayers, with reduced Tc but largely enhanced upper critical field. Our study opens up the opportunity to study the interplay between quantum confinement, interfacial symmetry breaking and superconductivity in an epitaxial silicide film, which is technologically relevant in microelectronics.
This paper is concerned with self-supervised learning for small models. The problem is motivated by our empirical studies that while the widely used contrastive self-supervised learning method has shown great progress on large model training, it does
not work well for small models. To address this problem, we propose a new learning paradigm, named SElf-SupErvised Distillation (SEED), where we leverage a larger network (as Teacher) to transfer its representational knowledge into a smaller architecture (as Student) in a self-supervised fashion. Instead of directly learning from unlabeled data, we train a student encoder to mimic the similarity score distribution inferred by a teacher over a set of instances. We show that SEED dramatically boosts the performance of small networks on downstream tasks. Compared with self-supervised baselines, SEED improves the top-1 accuracy from 42.2% to 67.6% on EfficientNet-B0 and from 36.3% to 68.2% on MobileNet-v3-Large on the ImageNet-1k dataset.
Flying plasma mirrors induced by intense lasers has been proposed as a promising way to generate few-cycle EUV or X-ray lasers. In addition, if such a relativistic plasma mirror can accelerate, then it would serve as an analog black hole to investiga
te the information loss paradox associated with the black hole Hawking evaporation. Among these applications, the reflectivity, which is usually frequency-dependent, would affect the outgoing photon spectrum and therefore impact on the analysis of the physics under investigation. In this paper, these two issues are investigated analytically and numerically with one-dimensional particle-in-cell (PIC) simulations. Based on our simulation results, we propose a new model that provides a better estimate of the reflectivity than those studied previously. Besides, we found that the peak frequency of the reflected spectrum of a gaussian incident wave deviates from the expected value, $4gamma^2omega$, due to the dependence of reflectivity on the frequency of the incident wave.
