Searcharxiv⌕ Search

arXiv subjects

Liang Dong

Publications and source records attributed to Liang Dong.

At least 55 records · Page 3Linked to original sources
arXiv2025

GaussianCAD: Robust Self-Supervised CAD Reconstruction from Three Orthographic Views Using 3D Gaussian Splatting

The automatic reconstruction of 3D computer-aided design (CAD) models from CAD sketches has recently gained significant attention in the computer vision community. Most existing methods, however, rely on vector CAD sketches and 3D ground truth for supervision, which are often difficult to be obtained in industrial applications and are sensitive to noise inputs. We propose viewing CAD reconstruction as a specific instance of sparse-view 3D reconstruction to overcome these limitations. While this reformulation offers a promising perspective, existing 3D reconstruction methods typically require natural images and corresponding camera poses as inputs, which introduces two major significant challenges: (1) modality discrepancy between CAD sketches and natural images, and (2) difficulty of accurate camera pose estimation for CAD sketches. To solve these issues, we first transform the CAD sketches into representations resembling natural images and extract corresponding masks. Next, we manually calculate the camera poses for the orthographic views to ensure accurate alignment within the 3D coordinate system. Finally, we employ a customized sparse-view 3D reconstruction method to achieve high-quality reconstructions from aligned orthographic views. By leveraging raster CAD sketches for self-supervision, our approach eliminates the reliance on vector CAD sketches and 3D ground truth. Experiments on the Sub-Fusion360 dataset demonstrate that our proposed method significantly outperforms previous approaches in CAD reconstruction performance and exhibits strong robustness to noisy inputs.

cs.CV↗
arXiv2025

Multiwavelength Variability Analysis of the Blazar PKS 0727-11: A $\sim$168 Days Quasi-periodic Oscillation in Gamma-ray

We performed variability analysis of the multiwavelength light curves for the flat-spectrum radio quasar PKS 0727-11. Using the generalized Lomb-Scargle periodogram, we identified a possible quasi-periodic oscillation (QPO) of $\sim$ 168.6 days (persisted for 6 cycles, with a significance of $3.8σ$) in the gamma-ray light curve during the flare period (MJD 54687-55738). It is the first time that periodic variations have been detected in this source, and further supported by other methods: weighted wavelet $z$-transform, phase dispersion minimization, REDFIT, autoregressive integrated moving average model, and structure function analysis. Cross-correlation analysis shows that there is a strong correlation between multi-band light variations, indicating that gamma-ray and radio flares may originate from the same disturbance, and the distance between the emission regions of gamma-ray and radio flares is calculated based on the time lag. We demonstrate that QPO arising from the non-ballistic helical jet motion driven by the orbital motion in a supermassive binary black hole is a plausible physical explanation. In this scenario, the estimated mass of the primary black hole is $M\sim3.66\times10^8-5.79\times10^{9}M_\odot$.

astro-ph.HE↗
arXiv2025

Two repeated quasi-periodic oscillations in the FSRQ S5 1044+71 observed by TESS

In this work, we report for the first time two repeated quasi-periodic oscillations (QPOs) in the light curve of the Flat Spectrum Radio Quasar (FSRQ) S5 1044+71. This source was observed by the Transiting Exoplanet Survey Satellite (TESS) in multiple sectors. We used the generalized Lomb-Scargle periodogram method and weighted wavelet Z-transform method to search for significant periodic signals. The main results are as follows: We found QPOs of $\sim$ 7.0 days (persisted for 4 cycles, with a significance of $\sim3.5σ$) and $\sim$ 7.3 days (persisted for 5 cycles, with a significance of $\sim3.8σ$) in the light curves of Sector 47 and EP1, respectively. Considering range of error, we consider them to be the same. We discussed two likely models of these rapid quasi-periodic variations: One comes from the jet and the other from the accretion disk. For the first one, we consider kink instability of the jet as a plausible explanation. Second, the QPO is probable to come from the main hot spots in the accretion disk, which is located approximately within the innermost stable circular orbit allowed by general relativity. Based on this model, we estimate the mass of the black hole in S5 1044+71 to be $3.49 \times 10^9 M_{\odot}$.

astro-ph.HE↗
arXiv2024

Persistent but weak magnetic field at Moon's midlife revealed by Chang'e-5 basalt

The evolution of the lunar magnetic field can reveal the Moon's interior structure, thermal history, and surface environment. The mid-to-late stage evolution of the lunar magnetic field is poorly constrained, and thus the existence of a long-lived lunar dynamo remains controversial. The Chang'e-5 mission returned the heretofore youngest mare basalts from Oceanus Procellarum uniquely positioned at mid-latitude. We recovered weak paleointensities of 2-4 uT from the Chang'e-5 basalt clasts at 2 billion years ago, attestting to the longevity of a lunar dynamo until at least the Moon's midlife. This paleomagnetic result implies the existence of thermal convection in the lunar deep interior at the lunar mid-stage which may have supplied mantle heat flux for the young volcanism.

physics.geo-ph↗
arXiv2024

Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent

In this paper, we introduce Hunyuan-Large, which is currently the largest open-source Transformer-based mixture of experts model, with a total of 389 billion parameters and 52 billion activation parameters, capable of handling up to 256K tokens. We conduct a thorough evaluation of Hunyuan-Large's superior performance across various benchmarks including language understanding and generation, logical reasoning, mathematical problem-solving, coding, long-context, and aggregated tasks, where it outperforms LLama3.1-70B and exhibits comparable performance when compared to the significantly larger LLama3.1-405B model. Key practice of Hunyuan-Large include large-scale synthetic data that is orders larger than in previous literature, a mixed expert routing strategy, a key-value cache compression technique, and an expert-specific learning rate strategy. Additionally, we also investigate the scaling laws and learning rate schedule of mixture of experts models, providing valuable insights and guidances for future model development and optimization. The code and checkpoints of Hunyuan-Large are released to facilitate future innovations and applications. Codes: https://github.com/Tencent/Hunyuan-Large Models: https://huggingface.co/tencent/Tencent-Hunyuan-Large

cs.CL↗
arXiv2024

A Novel Hybrid Parameter-Efficient Fine-Tuning Approach for Hippocampus Segmentation and Alzheimer's Disease Diagnosis

Deep learning methods have significantly advanced medical image segmentation, yet their success hinges on large volumes of manually annotated data, which require specialized expertise for accurate labeling. Additionally, these methods often demand substantial computational resources, particularly for three-dimensional medical imaging tasks. Consequently, applying deep learning techniques for medical image segmentation with limited annotated data and computational resources remains a critical challenge. In this paper, we propose a novel parameter-efficient fine-tuning strategy, termed HyPS, which employs a hybrid parallel and serial architecture. HyPS updates a minimal subset of model parameters, thereby retaining the pre-trained model's original knowledge tructure while enhancing its ability to learn specific features relevant to downstream tasks. We apply this strategy to the state-of-the-art SwinUNETR model for medical image segmentation. Initially, the model is pre-trained on the BraTs2021 dataset, after which the HyPS method is employed to transfer it to three distinct hippocampus datasets.Extensive experiments demonstrate that HyPS outperforms baseline methods, especially in scenarios with limited training samples. Furthermore, based on the segmentation results, we calculated the hippocampal volumes of subjects from the ADNI dataset and combined these with metadata to classify disease types. In distinguishing Alzheimer's disease (AD) from cognitively normal (CN) individuals, as well as early mild cognitive impairment (EMCI) from late mild cognitive impairment (LMCI), HyPS achieved classification accuracies of 83.78% and 64.29%, respectively. These findings indicate that the HyPS method not only facilitates effective hippocampal segmentation using pre-trained models but also holds potential for aiding Alzheimer's disease detection. Our code is publicly available.

eess.IV↗
arXiv2024

Light-induced fictitious magnetic fields for quantum storage in cold atomic ensembles

In this work, we have demonstrated that optically generated fictitious magnetic fields can be utilized to extend the lifetime of quantum memories in cold atomic ensembles. All the degrees of freedom of an AC Stark shift such as polarization, spatial profile, and temporal waveform can be readily controlled in a precise manner. Temporal fluctuations over several experimental cycles, and spatial inhomogeneities along a cold atomic gas have been compensated by an optical beam. The advantage of the use of fictitious magnetic fields for quantum storage stems from the speed and spatial precision that these fields can be synthesized. Our simple and versatile technique can find widespread application in coherent pulse and single-photon storage in any atomic species.

quant-ph↗
arXiv2024

NEAT: Distilling 3D Wireframes from Neural Attraction Fields

This paper studies the problem of structured 3D reconstruction using wireframes that consist of line segments and junctions, focusing on the computation of structured boundary geometries of scenes. Instead of leveraging matching-based solutions from 2D wireframes (or line segments) for 3D wireframe reconstruction as done in prior arts, we present NEAT, a rendering-distilling formulation using neural fields to represent 3D line segments with 2D observations, and bipartite matching for perceiving and distilling of a sparse set of 3D global junctions. The proposed {NEAT} enjoys the joint optimization of the neural fields and the global junctions from scratch, using view-dependent 2D observations without precomputed cross-view feature matching. Comprehensive experiments on the DTU and BlendedMVS datasets demonstrate our NEAT's superiority over state-of-the-art alternatives for 3D wireframe reconstruction. Moreover, the distilled 3D global junctions by NEAT, are a better initialization than SfM points, for the recently-emerged 3D Gaussian Splatting for high-fidelity novel view synthesis using about 20 times fewer initial 3D points. Project page: \url{https://xuenan.net/neat}.

cs.CV↗
arXiv2023

Self-supervised Deep Unrolled Reconstruction Using Regularization by Denoising

Deep learning methods have been successfully used in various computer vision tasks. Inspired by that success, deep learning has been explored in magnetic resonance imaging (MRI) reconstruction. In particular, integrating deep learning and model-based optimization methods has shown considerable advantages. However, a large amount of labeled training data is typically needed for high reconstruction quality, which is challenging for some MRI applications. In this paper, we propose a novel reconstruction method, named DURED-Net, that enables interpretable self-supervised learning for MR image reconstruction by combining a self-supervised denoising network and a plug-and-play method. We aim to boost the reconstruction performance of Noise2Noise in MR reconstruction by adding an explicit prior that utilizes imaging physics. Specifically, the leverage of a denoising network for MRI reconstruction is achieved using Regularization by Denoising (RED). Experiment results demonstrate that the proposed method requires a reduced amount of training data to achieve high reconstruction quality among the state-of-art of MR reconstruction utilizing the Noise2Noise method.

eess.IV↗
arXiv2023

Patched Line Segment Learning for Vector Road Mapping

This paper presents a novel approach to computing vector road maps from satellite remotely sensed images, building upon a well-defined Patched Line Segment (PaLiS) representation for road graphs that holds geometric significance. Unlike prevailing methods that derive road vector representations from satellite images using binary masks or keypoints, our method employs line segments. These segments not only convey road locations but also capture their orientations, making them a robust choice for representation. More precisely, given an input image, we divide it into non-overlapping patches and predict a suitable line segment within each patch. This strategy enables us to capture spatial and structural cues from these patch-based line segments, simplifying the process of constructing the road network graph without the necessity of additional neural networks for connectivity. In our experiments, we demonstrate how an effective representation of a road graph significantly enhances the performance of vector road mapping on established benchmarks, without requiring extensive modifications to the neural network architecture. Furthermore, our method achieves state-of-the-art performance with just 6 GPU hours of training, leading to a substantial 32-fold reduction in training costs in terms of GPU hours.

cs.CV↗
arXiv2023

Comprehensive study of the blazars from Fermi-LAT LCR: The log-normal flux distribution and linear RMS-Flux relation

Fermi-LAT LCR provide continuous and regularly-sampled gamma-ray light curves, spanning about 14 years, for a large sample of blazars. The log-normal flux distribution and linear RMS-Flux relation of the light curves for a few of Fermi blazar have been examined in previous studies. However, the probability that blazars exhibit log-normal flux distribution and linear RMS-Flux relation in their gamma-ray light curves has not been systematically explored. In this study, we comprehensively research on the distribution of gamma-ray flux and the statistical characteristics on a large sample of 1414 variable blazars from the Fermi-LAT LCR catalog, including 572 FSRQs, 477 BL Lacs, and 365 BCUs, and statistically compare their flux distributions with normal and log-normal distributions. The results indicate that the probability of not reject log-normal is 42.05% for the large sample, and there is still 2.05% probability of not reject normality, based on the joint of Kolmogorov-Smirnov, Shapiro-Wilk and Normality tests. We further find that the probability that BL Lacs conforms to the log-normal distribution is higher than that of FSRQs. Besides, after removing sources with less than 200 data points from this large sample, a sample of 549 blazars, which is still a large sample comparing to the previous studies, was obtained. Basing on dividing the light curves into segments every 20 points (or 40 points, or one year), we fitted the linear RMS-Flux relation of this three different sets, and found that the Pearson correlation coefficients are all close to 1 of the most blazars. This result indicates a strong linear correlation between the RMS and the flux of this 549 blazars. The log-normal distribution and linear RMS-Flux relation indicate that the variability of gamma-ray flux for most blazars is non-linear and multiplicative process.

astro-ph.HE↗
arXiv2022

Complete two-loop QCD amplitudes for $tW$ production at hadron colliders

We calculate the complete two-loop QCD amplitudes for hadronic $tW$ production by combining analytical and numerical techniques. The amplitudes have been first reduced to master integrals of eight planar and seven non-planar families, which can contain at most four massive propagators. Then a rational transformation of the master integrals is found to obtain a good basis so that the dimensional parameter decouples from the kinematic variables in the denominators of reduction coefficients. The master integrals are computed by solving their differential equations numerically. We find that the finite part of the two-loop squared amplitude is stable in the bulk of the phase space. After phase space integration and convolution with the parton distributions, it increases the LO cross section by about $3\%$.

hep-ph↗
arXiv2022

One-loop squared amplitudes for hadronic $tW$ production at next-to-next-to-leading order in QCD

We present the analytic results of one-loop squared amplitudes for $tW$ production at a hadron collider. The calculation is performed using the method of differential equations. After renormalization, we have checked that the infrared divergences agree with the general structure predicted by anomalous dimensions. The finite remainder contributes to the next-to-next-to-leading order hard function, one of the essential ingredients in the factorization formula of the cross section near the infrared region, which can be used in resummation of all-order soft gluon effects or a differential next-to-next-to-leading order calculation based on the phase space slicing method.

hep-ph↗
arXiv2022

Analytic two-loop amplitudes for $tW$ production: leading color and light fermion-loop contributions

We present the analytical results of the two-loop amplitudes for hadronic $tW$ production, focusing on the leading color and light fermion-loop contributions. The calculation of the two-loop integrals is performed using the method of canonical differential equations. The results have been expressed in terms of multiple polylogarithms and checked by comparing the infra-red divergences with the predictions from anomalous dimensions. Combined with the one-loop squared amplitudes we have computed previously, we obtain the hard function relevant to a NNLO Monte Carlo calculation. We find that the hard function varies slowly in the region with small top quark velocity but increases dramatically in the region with very large top quark velocity. After phase space integration, the leading color hard function gives an about $5.4\%$ correction to the leading order cross section, while the light fermion loop contributes about $-1.4\%$.

hep-ph↗
arXiv2021

Turbulence-based load alleviation control for wind turbine in extreme turbulence situation

The extreme loads experienced by the wind turbine in the extreme wind events are critical for the evaluation of structural reliability. Hence, the load alleviation control methods need to be designed and deployed to reduce the adverse effects of extreme wind events. This work demonstrates that the extreme loads are highly correlated to wind conditions such as turbulence-induced wind shears. Based on this insight, this work proposes a turbulence-based load alleviation control strategy for adapting the controller to changes in wind condition. The estimation of the rotor averaged wind shear based on the rotor loads is illustrated, and is herein used to statistically characterize the extreme wind events for control purpose. To demonstrates the benefits, simulations are carried out using high-fidelity aero-elastic tool and the DTU 10 MW reference turbine in normal and extreme turbulence wind conditions. The results indicate that the proposed method can effectively decrease the exceedance probability of the extreme loads. Meanwhile, the method can minimize the loss of annual energy production in normal operating condition.

eess.SY↗
arXiv2020

Berry curvature effects on quasiparticle dynamics in superconductors

We construct a theory for the semiclassical dynamics of superconducting quasiparticles by following their wave-packet motion and reveal rich contents of Berry curvature effects in the phase-space spanned by position and momentum. These Berry curvatures are traced back to the characteristics of superconductivity, including the nontrivial momentum-space geometry of superconducting pairing, the real-space supercurrent, and the charge dipole of quasiparticles. The Berry-curvature effects strongly influence the spectroscopic and transport properties of superconductors, such as the local density of states and the thermal Hall conductivity. As a model illustration, we apply the theory to study the twisted bilayer graphene with a $d_{x^{2}+y^{2}}+id_{xy}$ superconducting gap function, and demonstrate Berry-curvature induced effects.

cond-mat.supr-con↗
arXiv2020

Conceptualized Representation Learning for Chinese Biomedical Text Mining

Biomedical text mining is becoming increasingly important as the number of biomedical documents and web data rapidly grows. Recently, word representation models such as BERT has gained popularity among researchers. However, it is difficult to estimate their performance on datasets containing biomedical texts as the word distributions of general and biomedical corpora are quite different. Moreover, the medical domain has long-tail concepts and terminologies that are difficult to be learned via language models. For the Chinese biomedical text, it is more difficult due to its complex structure and the variety of phrase combinations. In this paper, we investigate how the recently introduced pre-trained language model BERT can be adapted for Chinese biomedical corpora and propose a novel conceptualized representation learning approach. We also release a new Chinese Biomedical Language Understanding Evaluation benchmark (\textbf{ChineseBLUE}). We examine the effectiveness of Chinese pre-trained models: BERT, BERT-wwm, RoBERTa, and our approach. Experimental results on the benchmark show that our approach could bring significant gain. We release the pre-trained model on GitHub: https://github.com/alibaba-research/ChineseBLUE.

cs.CL↗