arXiv subjects

Xue Rui

Publications and source records attributed to Xue Rui.

9 recordsLinked to original sources

arXiv2026

Trustworthy Data-Driven Wildfire Risk Prediction and Understanding in Western Canada

In recent decades, the intensification of wildfire activity in western Canada has resulted in substantial socio-economic and environmental losses. Accurate wildfire risk prediction is hindered by the intrinsic stochasticity of ignition and spread and by nonlinear interactions among fuel conditions, meteorology, climate variability, topography, and human activities, challenging the reliability and interpretability of purely data-driven models. We propose a trustworthy data-driven wildfire risk prediction framework based on long-sequence, multi-scale temporal modeling, which integrates heterogeneous drivers while explicitly quantifying predictive uncertainty and enabling process-level interpretation. Evaluated over western Canada during the record-breaking 2023 and 2024 fire seasons, the proposed model outperforms existing time-series approaches, achieving an F1 score of 0.90 and a PR-AUC of 0.98 with low computational cost. Uncertainty-aware analysis reveals structured spatial and seasonal patterns in predictive confidence, highlighting increased uncertainty associated with ambiguous predictions and spatiotemporal decision boundaries. SHAP-based interpretation provides mechanistic understanding of wildfire controls, showing that temperature-related drivers dominate wildfire risk in both years, while moisture-related constraints play a stronger role in shaping spatial and land-cover-specific contrasts in 2024 compared to the widespread hot and dry conditions of 2023. Data and code are available at https://github.com/SynUW/mmFire.

cs.CV↗

arXiv2025

Leptohadronic Multimessenger Modeling of Two High-redshift (z $>$ 1) Neutrino Emission Blazar Candidates

The blazars are one of the leading candidate sources of high-energy neutrinos. Recently, two blazars have been found to be temporally and spatially correlated with some IceCube high-energy neutrino events. The two blazars, GB6 J2113+1121 and NVSS J171822+423948, are Flat Spectrum Radio Quasars (FSRQs) with redshifts greater than unity. In particular, NVSS J171822+423948 has a redshift of 2.7, which provides an important probe for studying the radiation processes of jets from active galactic nuclei in the early universe. To better understand the physical origin of the IceCube neutrinos, we adopt the one-zone leptohadronic model to fit the multimessenger emission of GB6 J2113+1121 and NVSS J171822+423948 during their $\gamma$-ray flaring periods and then calculate the high-energy neutrino detection probability. The chance of detecting a single muon neutrino from these two sources is found to be $\sim 2\%$ and $0.8\%$, respectively. Although such detection rates are not high mainly because of their high redshifts, our investigation strongly suggests that these sources are efficient PeV neutrino emitters. Our results also indicate that electromagnetic cascades produced by hadronic processes contribute significantly to X-ray and $\gamma$-ray emissions. However, high-energy $\gamma$-rays can be severely absorbed by the soft photon field from the broad-line region (BLR), which weakens the correlation between $\gamma$-rays and neutrinos, while suggesting a stronger connection between X-rays and neutrinos. We predict that IceCube will continue to detect neutrinos from FSRQs with redshifts greater than 1 in the future.

astro-ph.HE↗

arXiv2024

One-shot Generative Domain Adaptation in 3D GANs

3D-aware image generation necessitates extensive training data to ensure stable training and mitigate the risk of overfitting. This paper first considers a novel task known as One-shot 3D Generative Domain Adaptation (GDA), aimed at transferring a pre-trained 3D generator from one domain to a new one, relying solely on a single reference image. One-shot 3D GDA is characterized by the pursuit of specific attributes, namely, high fidelity, large diversity, cross-domain consistency, and multi-view consistency. Within this paper, we introduce 3D-Adapter, the first one-shot 3D GDA method, for diverse and faithful generation. Our approach begins by judiciously selecting a restricted weight set for fine-tuning, and subsequently leverages four advanced loss functions to facilitate adaptation. An efficient progressive fine-tuning strategy is also implemented to enhance the adaptation process. The synergy of these three technological components empowers 3D-Adapter to achieve remarkable performance, substantiated both quantitatively and qualitatively, across all desired properties of 3D GDA. Furthermore, 3D-Adapter seamlessly extends its capabilities to zero-shot scenarios, and preserves the potential for crucial tasks such as interpolation, reconstruction, and editing within the latent space of the pre-trained generator. Code will be available at https://github.com/iceli1007/3D-Adapter.

cs.CV↗

arXiv2024

Deep Learning for Wildfire Risk Prediction: Integrating Remote Sensing and Environmental Data

Wildfires pose a significant threat to ecosystems, wildlife, and human communities, leading to habitat destruction, pollutant emissions, and biodiversity loss. Accurate wildfire risk prediction is crucial for mitigating these impacts and safeguarding both environmental and human health. This paper provides a comprehensive review of wildfire risk prediction methodologies, with a particular focus on deep learning approaches combined with remote sensing. We begin by defining wildfire risk and summarizing the geographical distribution of related studies. In terms of data, we analyze key predictive features, including fuel characteristics, meteorological and climatic conditions, socioeconomic factors, topography, and hydrology, while also reviewing publicly available wildfire prediction datasets derived from remote sensing. Additionally, we emphasize the importance of feature collinearity assessment and model interpretability to improve the understanding of prediction outcomes. Regarding methodology, we classify deep learning models into three primary categories: time-series forecasting, image segmentation, and spatiotemporal prediction, and further discuss methods for converting model outputs into risk classifications or probability-adjusted predictions. Finally, we identify the key challenges and limitations of current wildfire-risk prediction models and outline several research opportunities. These include integrating diverse remote sensing data, developing multimodal models, designing more computationally efficient architectures, and incorporating cross-disciplinary methods--such as coupling with numerical weather-prediction models--to enhance the accuracy and robustness of wildfire-risk assessments.

cs.LG↗

arXiv2023

Peer is Your Pillar: A Data-unbalanced Conditional GANs for Few-shot Image Generation

Few-shot image generation aims to train generative models using a small number of training images. When there are few images available for training (e.g. 10 images), Learning From Scratch (LFS) methods often generate images that closely resemble the training data while Transfer Learning (TL) methods try to improve performance by leveraging prior knowledge from GANs pre-trained on large-scale datasets. However, current TL methods may not allow for sufficient control over the degree of knowledge preservation from the source model, making them unsuitable for setups where the source and target domains are not closely related. To address this, we propose a novel pipeline called Peer is your Pillar (PIP), which combines a target few-shot dataset with a peer dataset to create a data-unbalanced conditional generation. Our approach includes a class embedding method that separates the class space from the latent space, and we use a direction loss based on pre-trained CLIP to improve image diversity. Experiments on various few-shot datasets demonstrate the advancement of the proposed PIP, especially reduces the training requirements of few-shot image generation.

cs.CV↗

arXiv2023

A Proxy Attack-Free Strategy for Practically Improving the Poisoning Efficiency in Backdoor Attacks

Poisoning efficiency is crucial in poisoning-based backdoor attacks, as attackers aim to minimize the number of poisoning samples while maximizing attack efficacy. Recent studies have sought to enhance poisoning efficiency by selecting effective samples. However, these studies typically rely on a proxy backdoor injection task to identify an efficient set of poisoning samples. This proxy attack-based approach can lead to performance degradation if the proxy attack settings differ from those of the actual victims, due to the shortcut nature of backdoor learning. Furthermore, proxy attack-based methods are extremely time-consuming, as they require numerous complete backdoor injection processes for sample selection. To address these concerns, we present a Proxy attack-Free Strategy (PFS) designed to identify efficient poisoning samples based on the similarity between clean samples and their corresponding poisoning samples, as well as the diversity of the poisoning set. The proposed PFS is motivated by the observation that selecting samples with high similarity between clean and corresponding poisoning samples results in significantly higher attack success rates compared to using samples with low similarity. Additionally, we provide theoretical foundations to explain the proposed PFS. We comprehensively evaluate the proposed strategy across various datasets, triggers, poisoning rates, architectures, and training hyperparameters. Our experimental results demonstrate that PFS enhances backdoor attack efficiency while also offering a remarkable speed advantage over previous proxy attack-based selection methodologies.

cs.CR↗

arXiv2021

Exploring The Effect of High-frequency Components in GANs Training

Generative Adversarial Networks (GANs) have the ability to generate images that are visually indistinguishable from real images. However, recent studies have revealed that generated and real images share significant differences in the frequency domain. In this paper, we explore the effect of high-frequency components in GANs training. According to our observation, during the training of most GANs, severe high-frequency differences make the discriminator focus on high-frequency components excessively, which hinders the generator from fitting the low-frequency components that are important for learning images' content. Then, we propose two simple yet effective frequency operations for eliminating the side effects caused by high-frequency differences in GANs training: High-Frequency Confusion (HFC) and High-Frequency Filter (HFF). The proposed operations are general and can be applied to most existing GANs with a fraction of the cost. The advanced performance of the proposed operations is verified on multiple loss functions, network architectures, and datasets. Specifically, the proposed HFF achieves significant improvements of $42.5\%$ FID on CelebA (128*128) unconditional generation based on SNGAN, $30.2\%$ FID on CelebA unconditional generation based on SSGAN, and $69.3\%$ FID on CelebA unconditional generation based on InfoMAXGAN.

cs.CV↗

arXiv2018

Atomic-resolution study of oxygen vacancy ordering in $La_{0.5}$$Sr_{0.5}$Co$O_{3-δ}$ thin films on SrTi$O_{3}$ during in-situ cooling experiments

The presence of oxygen vacancy, as well as ordering of vacancies plays an important role in determining the electronic, ionic and thermal transport properties of many transition metal oxide materials. Controlling the concentration of oxygen vacancies as well as the structures or domains of ordered oxygen vacancies has been the subject of many experimental and theoretical studies. In epitaxial thin films, the concentration of oxygen vacancies as well as the type of ordering depends on the structure of the support as well as the lattice mismatch between the thin films and the support. The role of temperature induced structural phase transitions on the oxygen vacancy ordering has remained largely unexplored. Here, we use aberration-corrected scanning transmission electron microscopy (STEM) combined with an in-situ cooling experiments to characterize the atomic/electronic structures of oxygen-deficient $La_{0.5}$$Sr_{0.5}$Co$O_{3-δ}$ thin films grown on SrTi$O_{3}$ across the anti-ferrodistortive phase transition of SrTi$O_{3}$ at 105 K. We demonstrate that atomic-resolution imaging and electron energy-loss spectroscopy (EELS) can be used to examine variations in the local density of states as a function of sample temperature and thus of the structure of the support.

cond-mat.mes-hall↗

arXiv2016

Experimental verification of orbital engineering at the atomic scale: charge transfer and symmetry breaking in nickelate heterostructures

Epitaxial strain, layer confinement and inversion symmetry breaking have emerged as powerful new approaches to control the electronic and atomic-scale structural properties in complex metal oxides. Nickelate heterostructures, based on RENiO$_3$, where RE is a trivalent rare-earth cation, have been shown to be relevant model systems since the orbital occupancy, degeneracy, and, consequently, the electronic/magnetic properties can be altered as a function of epitaxial strain, layer thickness and superlattice structure. One such recent example is the tri-component LaTiO$_3$-LaNiO$_3$-LaAlO$_3$ superlattice, which exhibits charge transfer and orbital polarization as the result of its interfacial dipole electric field. A crucial step towards control of these parameters for future electronic and magnetic device applications is to develop an understanding of both the magnitude and range of the octahedral network's response towards interfacial strain and electric fields. An approach that provides atomic-scale resolution and sensitivity towards the local octahedral distortions and orbital occupancy is therefore required. Here, we employ atomic-resolution imaging coupled with electron spectroscopies and first principles theory to examine the role of interfacial charge transfer and symmetry breaking in a tricomponent nickelate superlattice system. We find that nearly complete charge transfer occurs between the LaTiO$_3$ and LaNiO$_3$ layers, resulting in a Ni$^{2+}$ valence state. We further demonstrate that this charge transfer is highly localized with a range of about 1 unit cell, within the LaNiO$_3$ layers. The results presented here provide important feedback to synthesis efforts aimed at stabilizing new electronic phases that are not accessible by conventional bulk or epitaxial film approaches.

cond-mat.mtrl-sci↗