arXiv subjects

Ke Lv

Publications and source records attributed to Ke Lv.

7 recordsLinked to original sources

arXiv2026

Direct Satellite-to-Device Communications: From Cooperative Task Offloading to Non-Cooperative Access Monitoring

Direct satellite-to-device (DS2D) communication is emerging as a transformative paradigm for extending ubiquitous connectivity and edge computing capabilities to remote and underserved regions within 6G non-terrestrial networks. However, practical deployment faces dual critical challenges: i) dynamic satellite channel conditions (e.g., severe Doppler shifts, fast fading) and constrained satellite computing resources in cooperative scenarios; and ii) unauthorized satellite access introduces significant spectrum security threats in non-cooperative scenarios. To address these challenges, we propose a versatile DS2D system that supports cooperative task offloading and non-cooperative access monitoring. For cooperative DS2D communications, we integrate a channel estimation module with a dueling double deep Q-network (D3QN) to dynamically optimize task offloading strategy. For non-cooperative DS2D communications, we propose Transformer-based models to enable blind signal detection and automatic modulation classification (AMC). Simulation results show that: 1) The D3QN algorithm reduces average latency by up to 225\% compared to static association policies. 2) Our signal detection model achieves an average presence detection probability of 90.5\% for DS2D signals. 3) The proposed AMC algorithm achieves superior performance across different signal-to-noise ratios (SNRs), with a 9.4\% accuracy gain in low-SNR regimes compared to existing methods.

cs.IT↗

arXiv2026

HeterSEED: Semantics-Structure Decoupling for Heterogeneous Graph Learning under Heterophily

Many real-world heterogeneous graphs exhibit pronounced heterophily, where connected nodes often have dissimilar labels or play different semantic roles. In such settings, standard heterogeneous graph neural networks that aggregate messages along metapaths or meta-relations primarily based on feature similarity can propagate misleading information, since feature similarity may be misaligned with underlying relational semantics. In this paper, we propose HeterSEED, a semantics-structure decoupling framework for heterogeneous graph learning under heterophily. HeterSEED decouples representation learning into a heterogeneous semantic channel that captures type- and relation-aware local semantics and a structure-aware heterophily channel that separates homophilic and heterophilic neighborhoods via pseudo-label-guided partitioning and aggregates them using metapath-based structural weights. A node-level adaptive fusion mechanism then combines the two channels to produce context-dependent node representations. Theoretically, we establish that, on heterogeneous graphs under heterophily, HeterSEED is strictly more expressive than standard heterogeneous graph neural networks that rely primarily on feature similarity and provably reduces the prediction bias introduced by heterophilic neighbors. Experiments on five real-world heterogeneous graphs, including two large-scale networks at the million-node and hundred-million-edge scale, demonstrate that HeterSEED consistently outperforms representative heterogeneous graph neural networks and recent heterophily-aware baselines, especially in strongly heterophilic regimes.

cs.LG↗

arXiv2025

Microscopic signatures of Chern number sign reversal in twisted bilayer WSe2

The discovery of quantized Chern numbers in twisted transition metal dichalcogenide (TMD) homobilayers,including 3.7{deg} twisted MoTe2 and 1.23{deg} twisted WSe2,has emerged as a defining breakthrough in physics. A striking and unresolved puzzle from these studies is the unexpected opposite sign of the observed Chern numbers between the two systems. Recent theory has proposed a twist-angle-dependent Chern number sign reversal in both twisted MoTe2 and WSe2, offering a potential explanation for the disparate experimental observations6. However, a direct experimental verification of the twist-angle-dependent Chern number sign reversal in a specific twisted TMD homobilayer is still elusive. Here, we report the first experimental demonstration that the Chern numbers of the moire frontier bands undergo sign reversal at a critical twist angle 1.42{deg} in twisted WSe2 bilayers (tWSe2). Using scanning tunnelling microscopy and spectroscopy, we direct measure layer-pseudospin skyrmion textures of tWSe2 and our results reveal that tWSe2 in the vicinity of the 1.42{deg} exhibits a twist-angle-dependent layer-pseudospin polarization, an effect that serves as the fundamental origin of the observed Chern number sigh reversal6-12.

cond-mat.mes-hall↗

arXiv2025

Building Lightweight Semantic Segmentation Models for Aerial Images Using Dual Relation Distillation

Recently, there have been significant improvements in the accuracy of CNN models for semantic segmentation. However, these models are often heavy and suffer from low inference speed, which limits their practical application. To address this issue, knowledge distillation has emerged as a promising approach to achieve a good trade-off between segmentation accuracy and efficiency. In this paper, we propose a novel dual relation distillation (DRD) technique that transfers both spatial and channel relations in feature maps from a cumbersome model (teacher) to a compact model (student). Specifically, we compute spatial and channel relation maps separately for the teacher and student models, and then align corresponding relation maps by minimizing their distance. Since the teacher model usually learns more information and collects richer spatial and channel correlations than the student model, transferring these correlations from the teacher to the student can help the student mimic the teacher better in terms of feature distribution, thus improving the segmentation accuracy of the student model. We conduct comprehensive experiments on three segmentation datasets, including two widely adopted benchmarks in the remote sensing field (Vaihingen and Potsdam datasets) and one popular benchmark in general scene (Cityscapes dataset). The experimental results demonstrate that our novel distillation framework can significantly boost the performance of the student network without incurring extra computational overhead.

eess.IV↗

arXiv2024

Edge-guided and Class-balanced Active Learning for Semantic Segmentation of Aerial Images

Semantic segmentation requires pixel-level annotation, which is time-consuming. Active Learning (AL) is a promising method for reducing data annotation costs. Due to the gap between aerial and natural images, the previous AL methods are not ideal, mainly caused by unreasonable labeling units and the neglect of class imbalance. Previous labeling units are based on images or regions, which does not consider the characteristics of segmentation tasks and aerial images, i.e., the segmentation network often makes mistakes in the edge region, and the edge of aerial images is often interlaced and irregular. Therefore, an edge-guided labeling unit is proposed and supplemented as the new unit. On the other hand, the class imbalance is severe, manifested in two aspects: the aerial image is seriously imbalanced, and the AL strategy does not fully consider the class balance. Both seriously affect the performance of AL in aerial images. We comprehensively ensure class balance from all steps that may occur imbalance, including initial labeled data, subsequent labeled data, and pseudo-labels. Through the two improvements, our method achieves more than 11.2\% gains compared to state-of-the-art methods on three benchmark datasets, Deepglobe, Potsdam, and Vaihingen, and more than 18.6\% gains compared to the baseline. Sufficient ablation studies show that every module is indispensable. Furthermore, we establish a fair and strong benchmark for future research on AL for aerial image segmentation.

cs.CV↗

arXiv2021

Spatial and Magnetic Confinement of Massless Dirac Fermions

The massless Dirac fermions and the ease to introduce spatial and magnetic confinement in graphene provide us unprecedented opportunity to explore confined relativistic matter in this condensed-matter system. Here we report the interplay between the confinement induced by external electric fields and magnetic fields of the massless Dirac fermions in graphene. When the magnetic length lB is larger than the characteristic length of the confined electric potential lV, the spatial confinement dominates and a relatively small critical magnetic field splits the spatial-confinement-induced atomic-like shell states by switching on a pi Berry phase of the quasiparticles. When the lB becomes smaller than the lV, the transition from spatial confinement to magnetic confinement occurs and the atomic-like shell states condense into Landau levels (LLs) of the Fock-Darwin states in graphene. Our experiment demonstrates that the spatial confinement dramatically changes the energy spacing between the LLs and generates large electron-hole asymmetry of the energy spacing between the LLs. These results shed light on puzzling observations in previous experiments, which hitherto remained unaddressed.

cond-mat.mes-hall↗

arXiv2019

The Richardson-Lucy Deconvolution method to Extract LAMOST 1D Spectra

We use the Richardson-Lucy deconvolution algorithm to extract one dimensional (1D) spectra from LAMOST spectrum images. Compared with other deconvolution algorithms, this algorithm is much more fast. The practice on a real LAMOST image illustrates that the 1D resulting spectrum of this method has a higher SNR and resolution than those extracted by the LAMOST pipeline. Furthermore, our algorithm can effectively depress the ringings that are often shown in the 1D resulting spectra of other deconvolution methods.

astro-ph.IM↗