Searcharxiv⌕ Search

arXiv subjects

Liang Zhang

Publications and source records attributed to Liang Zhang.

At least 271 records · Page 15Linked to original sources
arXiv2023

Adapted poling to break the nonlinear efficiency limit in nanophotonic lithium niobate waveguides

Nonlinear frequency mixing is of critical importance in extending the wavelength range of optical sources. It is also indispensable for emerging applications such as quantum information and photonic signal processing. Conventional lithium niobate with periodic poling is the most widely used device for frequency mixing due to the strong second-order nonlinearity. The recent development of nanophotonic lithium niobate waveguides promises improvements of nonlinear efficiencies by orders of magnitude with sub-wavelength optical conferment. However, the intrinsic nanoscale inhomogeneity in nanophotonic lithium niobate limits the coherent interaction length, leading to low nonlinear efficiencies. Therefore, the performance of nanophotonic lithium niobate waveguides is still far behind conventional counterparts. Here, we overcome this limitation and demonstrate ultra-efficient second order nonlinearity in nanophotonic lithium niobate waveguides significantly outperforming conventional crystals. This is realized by developing the adapted poling approach to eliminate the impact of nanoscale inhomogeneity in nanophotonic lithium niobate waveguides. We realize overall secondharmonic efficiency near 10^4 %/W without cavity enhancement, which saturates the theoretical limit. Phase-matching bandwidths and temperature tunability are improved through dispersion engineering. The ideal square dependence of the nonlinear efficiency on the waveguide length is recovered. We also break the trade-off between the energy conversion ratio and pump power. A conversion ratio over 80% is achieved in the single-pass configuration with pump power as low as 20 mW.

physics.optics↗
arXiv2023

FreeCOS: Self-Supervised Learning from Fractals and Unlabeled Images for Curvilinear Object Segmentation

Curvilinear object segmentation is critical for many applications. However, manually annotating curvilinear objects is very time-consuming and error-prone, yielding insufficiently available annotated datasets for existing supervised methods and domain adaptation methods. This paper proposes a self-supervised curvilinear object segmentation method that learns robust and distinctive features from fractals and unlabeled images (FreeCOS). The key contributions include a novel Fractal-FDA synthesis (FFS) module and a geometric information alignment (GIA) approach. FFS generates curvilinear structures based on the parametric Fractal L-system and integrates the generated structures into unlabeled images to obtain synthetic training images via Fourier Domain Adaptation. GIA reduces the intensity differences between the synthetic and unlabeled images by comparing the intensity order of a given pixel to the values of its nearby neighbors. Such image alignment can explicitly remove the dependency on absolute intensity values and enhance the inherent geometric characteristics which are common in both synthetic and real images. In addition, GIA aligns features of synthetic and real images via the prediction space adaptation loss (PSAL) and the curvilinear mask contrastive loss (CMCL). Extensive experimental results on four public datasets, i.e., XCAD, DRIVE, STARE and CrackTree demonstrate that our method outperforms the state-of-the-art unsupervised methods, self-supervised methods and traditional methods by a large margin. The source code of this work is available at https://github.com/TY-Shi/FreeCOS.

cs.CV↗
arXiv2023

A spectral-timing study of the inner flow geometry in MAXI J1535--571 with $Insight$-HXMT and NICER

We have performed a spectral-timing analysis on the black hole X-ray binary MAXI J1535--571 during its 2017 outburst, with the aim of exploring the evolution of the inner accretion flow geometry. X-ray reverberation lags are observed in the hard-intermediate state (HIMS) and soft-intermediate state (SIMS) of the outburst. During the HIMS, the characteristic frequency of the reverberation lags $ν_0$ (the frequency at which the soft lag turns to zero in the lag-frequency spectra) increases when the spectrum softens. This reflects a reduction of the spatial distance between the corona and accretion disc, when assuming the measured time lags are associated with the light travel time. We also find a strong correlation between $ν_0$ and type-C Quasi Periodic Oscillation (QPO) centroid frequency $ν_{QPO}$, which can be well explained by the Lense-Thirring (L-T) precession model under a truncated disk geometry. Despite the degeneracy in the spectral modellings, our results suggest that the accretion disc is largely truncated in the low hard state (LHS), and moves inward as the spectrum softens. Combine the spectral modelling results with the $ν_0$ - $ν_{QPO}$ evolution, we are inclined to believe that this source probably have a truncated disk geometry in the hard state.

astro-ph.HE↗
arXiv2023

The mHz quasi-regular modulations of 4U 1630--47 during its 1998 outburst

We present the results of a detailed timing and spectral analysis of the quasi-regular modulation (QRM) phenomenon in the black hole X-ray binary 4U 1630--47 during its 1998 outburst observed by Rossi X-ray Timing Explore (RXTE). We find that the $\sim$ 50-110 mHz QRM is flux dependent, and the QRM is detected with simultaneous low frequency quasi-periodic oscillations (LFQPOs). According to the behavior of the power density spectrum, we divide the observations into four groups. In the first group, namely behavior A, LFQPOs are detected, but no mHz QRM. The second group, namely behavior B, a QRM with frequency above $\sim$ 88 mHz is detected and the $\sim$ 5 Hz and $\sim$ 7 Hz LFQPOs are almost overlapping. In the third group, namely behavior C, the QRM frequency below $\sim$ 88 mHz is detected and the LFQPOs are significantly separated. In the forth group, namely behavior D, neither QRM nor LFQPOs are detected. We study the energy-dependence of the fractional rms, centroid frequency, and phase-lag of QRM and LFQPOs for behavior B and C. We then study the evolution of QRM and find that the frequency of QRM increases with hardness, while its rms decreases with hardness. We also analyze the spectra of each observation, and find that the QRM rms of behavior B has a positive correlation with $\rm F_{\rm powerlaw}$ / $\rm F_{\rm total}$. Finally, we give our understanding for this mHz QRM phenomena.

astro-ph.HE↗
arXiv2023

Movie101: A New Movie Understanding Benchmark

To help the visually impaired enjoy movies, automatic movie narrating systems are expected to narrate accurate, coherent, and role-aware plots when there are no speaking lines of actors. Existing works benchmark this challenge as a normal video captioning task via some simplifications, such as removing role names and evaluating narrations with ngram-based metrics, which makes it difficult for automatic systems to meet the needs of real application scenarios. To narrow this gap, we construct a large-scale Chinese movie benchmark, named Movie101. Closer to real scenarios, the Movie Clip Narrating (MCN) task in our benchmark asks models to generate role-aware narration paragraphs for complete movie clips where no actors are speaking. External knowledge, such as role information and movie genres, is also provided for better movie understanding. Besides, we propose a new metric called Movie Narration Score (MNScore) for movie narrating evaluation, which achieves the best correlation with human evaluation. Our benchmark also supports the Temporal Narration Grounding (TNG) task to investigate clip localization given text descriptions. For both two tasks, our proposed methods well leverage external knowledge and outperform carefully designed baselines. The dataset and codes are released at https://github.com/yuezih/Movie101.

cs.CV↗
arXiv2023

Anomalous Nernst effect induced terahertz emission in a single ferromagnetic film

By developing a bidirectional-pump terahertz (THz) emission spectroscopy, we reveal an anomalous Nernst effect (ANE) induced THz emission in a single ferromagnetic film. Based on the distinctive symmetry of the THz signals, ANE is unequivocally distinguished from the previously attributed ultrafast demagnetization and anomalous Hall effect mechanisms. A quantitative method is established to separate the different contributions, demonstrating a significant ANE contribution that even overwhelms other competing mechanisms. Our work not only clarifies the origin of the ferromagnetic-based THz emission, but also offers a fertile platform for investigating the ultrafast magnetism and THz spintronics.

cond-mat.mes-hall↗
arXiv2023

QPOML: A Machine Learning Approach to Detect and Characterize Quasi-Periodic Oscillations in X-ray Binaries

Astronomy is presently experiencing profound growth in the deployment of machine learning to explore large datasets. However, transient quasi-periodic oscillations (QPOs) which appear in power density spectra of many X-ray binary system observations are an intriguing phenomena heretofore not explored with machine learning. In light of this, we propose and experiment with novel methodologies for predicting the presence and properties of QPOs to make the first ever detections and characterizations of QPOs with machine learning models. We base our findings on raw energy spectra and processed features derived from energy spectra using an abundance of data from the NICER and RXTE space telescope archives for two black hole low mass X-ray binary sources, GRS 1915+105 and MAXI J1535-571. We advance these non-traditional methods as a foundation for using machine learning to discover global inter-object generalizations between - and provide unique insights about - energy and timing phenomena to assist with the ongoing challenge of unambiguously understanding the nature and origin of QPOs. Additionally, we have developed a publicly available Python machine learning library, QPOML, to enable further Machine Learning aided investigations into QPOs.

astro-ph.HE↗
arXiv2023

Integrated waveguide-based acousto-optic modulation with near-unity conversion efficiency

Acousto-optic modulation in piezoelectric materials offers the efficient method to bridge electrical and optical signals. It is widely used to control optical frequencies and intensities in modern optical systems including Q-switch lasers, ion traps, and optical tweezers. It is also critical for emerging applications such as quantum photonics and non-reciprocal optics. Acousto-optic devices have recently been demonstrated with promising performance on integrated platforms. However, the conversion efficiency of optical signals remains low in these integrated devices. This is attributed to the significant challenge in realizing large mode overlap, long interaction length, and high power robustness at the same time. Here, we develop acousto-optic devices with gallium nitride on sapphire substrate. The unique capability to confine both optical and acoustic fields in sub-wavelength scales without suspended structures allows efficient acousto-optic interactions over long distances under high driving power. This leads to the near-unity optical conversion efficiency with integrated acousto-optic modulators. With the unidirectional phase matching, we also demonstrate the non-reciprocal propagation of optical fields with isolation ratio above 10 dB. This work provides a robust and efficient acousto-optic platform, opening new opportunities for optical signal processing, quantum transduction, and non-magnetic optical isolation.

physics.optics↗
arXiv2023

Effect of solute atoms segregation on Al grain boundary properties by First-principles study

First-principles calculations were carried out to study the segregation behavior of Mg, and Cu and their effect on the energy and mechanical properties of different Al grain boundaries (GBs). Four symmetrical tilt GBs were selected for study, namely Σ5[001](210) GB, Σ5[001](310) GB, Σ9[110](221) GB, and Σ11[110](332) GB. The results show that both Mg and Cu have a segregation tendency at the GBs, and the segregation tendency of Cu is stronger than Mg. Mg is prone to form substitutional segregation at the GBs, but Cu is more likely to segregate at the interstitial sites. The segregation of Mg and Cu can reduce GB energy, and the GB energy continues to decrease with the increase of the segregation concentration. First-principles calculation tensile test shows that the segregation of Mg has a negative effect on the strength of GBs, and the GB strength decreases with the increase of the Mg concentration, while the GB strength was gradually enhanced with the increase of the Cu concentration. The strength of Σ5(210) GB and Σ9(221) GB are more sensitive to the segregation of solute atoms than the other two GBs. By calculating the charge density and the density of states of the pristine and the segregated GBs, it was found that the segregation of Mg caused charge depletion and structure expansion at the GBs, while the segregation of Cu increases the charge density of GBs and form new bonds with the surrounding Al atoms. The results provide useful information for improving the mechanical properties of materials by using the concept of GB segregation engineering.

cond-mat.mtrl-sci↗
arXiv2023

First-principles study of solute atoms segregation in Al Σ5(210) metastable grain boundaries

Grain boundary (GB) segregation of solute atoms plays an important role in the microstructure and macroscopic mechanical properties of materials. The study of GB segregation of solute atoms using computational simulation has become one of the hot spots in recent years. However, most studies mainly focus on ground-state GB structures with the lowest energy, and the impact of GB metastability with higher energy on solute segregation remains poorly understood. In this work, the first-principles method based on the density functional theory was adopted to investigate the effect of solute atoms Mg and Cu segregation on ground-state Σ5(210) GB (GB-I) and metastable GBs(GB-II, GB-III) in Al. GB energy, segregation energy, and theoretical tensile strength of Mg and Cu segregation at three GBs were calculated. The results show that both Mg and Cu have a large driving force to segregate to Al GBs, which reduces the GB energy and improves improve GB stability. The segregation of Mg and Cu on GB-III induces the transformation of the GB structural unit and the GB structural phase transformations. For the above three GBs, Cu segregation increases the theoretical tensile strength of GBs to varying degrees. The segregation of Mg would reduce the resistance of GB-I and GB-II, but enhances the strength of GB-III. The effect of solute atoms segregation on the mechanical properties of GBs was investigated by charge density distribution and density of states.

cond-mat.mtrl-sci↗
arXiv2023

Hilbert-Huang Transform analysis of quasi-periodic oscillations in MAXI J1820+070

We present time-frequency analysis, based on the Hilbert-Huang transform (HHT), of the evolution on the low-frequency quasi-periodic oscillations (LFQPOs) observed in the black hole X-ray binary MAXI J1820+070. Through the empirical mode decomposition (EMD) method, we decompose the light curve of the QPO component and measure its intrinsic phase lag between photons from different energy bands. We find that the QPO phase lag is negative (low energy photons lag behind high energy photons), meanwhile the absolute value of the lag increases with energy. By applying the Hilbert transform to the light curve of the QPO, we further extract the instantaneous frequency and amplitude of the QPO. Compared these results with those from the Fourier analysis, we find that the broadening of the QPO peak is mainly caused by the frequency modulation. Through further analysis, we find that these modulations could share a common physical origin with the broad-band noise, and can be well explained by the internal shock model of the jet.

astro-ph.HE↗
arXiv2023

InfoMetIC: An Informative Metric for Reference-free Image Caption Evaluation

Automatic image captioning evaluation is critical for benchmarking and promoting advances in image captioning research. Existing metrics only provide a single score to measure caption qualities, which are less explainable and informative. Instead, we humans can easily identify the problems of captions in details, e.g., which words are inaccurate and which salient objects are not described, and then rate the caption quality. To support such informative feedback, we propose an Informative Metric for Reference-free Image Caption evaluation (InfoMetIC). Given an image and a caption, InfoMetIC is able to report incorrect words and unmentioned image regions at fine-grained level, and also provide a text precision score, a vision recall score and an overall quality score at coarse-grained level. The coarse-grained score of InfoMetIC achieves significantly better correlation with human judgements than existing metrics on multiple benchmarks. We also construct a token-level evaluation dataset and demonstrate the effectiveness of InfoMetIC in fine-grained evaluation. Our code and datasets are publicly available at https://github.com/HAWLYQ/InfoMetIC.

cs.CV↗
arXiv2023

Biomarker Investigation using Multiple Brain Measures from MRI through XAI in Alzheimer's Disease Classification

Alzheimer's Disease (AD) is the world leading cause of dementia, a progressively impairing condition leading to high hospitalization rates and mortality. To optimize the diagnostic process, numerous efforts have been directed towards the development of deep learning approaches (DL) for the automatic AD classification. However, their typical black box outline has led to low trust and scarce usage within clinical frameworks. In this work, we propose two state-of-the art DL models, trained respectively on structural MRI (ResNet18) and brain connectivity matrixes (BC-GCN-SE) derived from diffusion data. The models were initially evaluated in terms of classification accuracy. Then, results were analyzed using an Explainable Artificial Intelligence (XAI) approach (Grad-CAM) to measure the level of interpretability of both models. The XAI assessment was conducted across 132 brain parcels, extracted from a combination of the Harvard-Oxford and AAL brain atlases, and compared to well-known pathological regions to measure adherence to domain knowledge. Results highlighted acceptable classification performance as compared to the existing literature (ResNet18: TPRmedian = 0.817, TNRmedian = 0.816; BC-GCN-SE: TPRmedian = 0.703, TNRmedian = 0.738). As evaluated through a statistical test (p < 0.05) and ranking of the most relevant parcels (first 15%), Grad-CAM revealed the involvement of target brain areas for both the ResNet18 and BC-GCN-SE models: the medial temporal lobe and the default mode network. The obtained interpretabilities were not without limitations. Nevertheless, results suggested that combining different imaging modalities may result in increased classification performance and model reliability. This could potentially boost the confidence laid in DL models and favor their wide applicability as aid diagnostic tools.

eess.IV↗
arXiv2023

An extreme ultraviolet wave associated with the possible expansion of sheared arcades

Context. Solar extreme ultraviolet (EUV) waves are propagating disturbances in the corona, and they usually accompany with various solar eruptions, from large-scale coronal mass ejections to small-scale coronal jets. Aims. Generally, it is believed that EUV waves are driven by the rapid expansion of coronal loops overlying the erupting cores. In this Letter, we present an exception of EUV wave that was not triggered by the expansion of coronal loops overlying the erupting core. Methods. Combining the multiwavelength observations from multiple instruments, we studied the event in detail. Results. The eruption was restricted in the active region (AR) and disturbed the nearby sheared arcades (SAs) connecting the source AR to a remote AR. Interestingly, following the disturbance, an EUV wave formed close to the SAs, but far away from the eruption source. Conclusions. All the results showed that the EUV wave had a closer temporal and spatial relationship with the disappearing part of SAs than the confined eruption. Hence, we suggest that the EUV wave was likely triggered by the expansion of some strands of SAs, rather than the expansion of erupting loops. It can be a possible complement for the driving mechanisms of EUV waves.

astro-ph.SR↗
arXiv2023

MPMQA: Multimodal Question Answering on Product Manuals

Visual contents, such as illustrations and images, play a big role in product manual understanding. Existing Product Manual Question Answering (PMQA) datasets tend to ignore visual contents and only retain textual parts. In this work, to emphasize the importance of multimodal contents, we propose a Multimodal Product Manual Question Answering (MPMQA) task. For each question, MPMQA requires the model not only to process multimodal contents but also to provide multimodal answers. To support MPMQA, a large-scale dataset PM209 is constructed with human annotations, which contains 209 product manuals from 27 well-known consumer electronic brands. Human annotations include 6 types of semantic regions for manual contents and 22,021 pairs of question and answer. Especially, each answer consists of a textual sentence and related visual regions from manuals. Taking into account the length of product manuals and the fact that a question is always related to a small number of pages, MPMQA can be naturally split into two subtasks: retrieving most related pages and then generating multimodal answers. We further propose a unified model that can perform these two subtasks all together and achieve comparable performance with multiple task-specific models. The PM209 dataset is available at https://github.com/AIM3-RUC/MPMQA.

cs.CL↗
arXiv2023

Road Network Representation Learning: A Dual Graph based Approach

Road network is a critical infrastructure powering many applications including transportation, mobility and logistics in real life. To leverage the input of a road network across these different applications, it is necessary to learn the representations of the roads in the form of vectors, which is named \emph{road network representation learning} (RNRL). While several models have been proposed for RNRL, they capture the pairwise relationships/connections among roads only (i.e., as a simple graph), and fail to capture among roads the high-order relationships (e.g., those roads that jointly form a local region usually have similar features such as speed limit) and long-range relationships (e.g., some roads that are far apart may have similar semantics such as being roads in residential areas). Motivated by this, we propose to construct a \emph{hypergraph}, where each hyperedge corresponds to a set of multiple roads forming a region. The constructed hypergraph would naturally capture the high-order relationships among roads with hyperedges. We then allow information propagation via both the edges in the simple graph and the hyperedges in the hypergraph in a graph neural network context. The graph reconstruction and hypergraph reconstruction tasks are conventional ones and can capture structural information. The hyperedge classification task can capture long-range relationships between pairs of roads that belong to hyperedges with the same label. We call the resulting model \emph{HyperRoad}. We further extend HyperRoad to problem settings when additional inputs of road attributes and/or trajectories that are generated on the roads are available.

cs.LG↗
arXiv2023

A Token-level Contrastive Framework for Sign Language Translation

Sign Language Translation (SLT) is a promising technology to bridge the communication gap between the deaf and the hearing people. Recently, researchers have adopted Neural Machine Translation (NMT) methods, which usually require large-scale corpus for training, to achieve SLT. However, the publicly available SLT corpus is very limited, which causes the collapse of the token representations and the inaccuracy of the generated tokens. To alleviate this issue, we propose ConSLT, a novel token-level \textbf{Con}trastive learning framework for \textbf{S}ign \textbf{L}anguage \textbf{T}ranslation , which learns effective token representations by incorporating token-level contrastive learning into the SLT decoding process. Concretely, ConSLT treats each token and its counterpart generated by different dropout masks as positive pairs during decoding, and then randomly samples $K$ tokens in the vocabulary that are not in the current sentence to construct negative examples. We conduct comprehensive experiments on two benchmarks (PHOENIX14T and CSL-Daily) for both end-to-end and cascaded settings. The experimental results demonstrate that ConSLT can achieve better translation quality than the strong baselines.

cs.CL↗
arXiv2023

Fast transitions of X-ray variability in the black hole transient GX 339--4: comparison with MAXI J1820+070 and MAXI J1348-630

Fast transitions between different types of power density spectra (PDS) happening over timescales of several tens of seconds are rare phenomena in black hole X-ray binaries. In this paper, we report a broadband spectral-timing analysis of the fast transitions observed in the 2021 outburst of GX 339-4 using NICER and HXMT observations. We observe transitions between band-limited noise-dominated PDS and type-B quasi-periodic oscillations (QPOs), and their rapid appearance or disappearance. We also make a detailed comparison between the fast transitions in GX 339-4 with those seen in MAXI J1820+070 and MAXI J1348--630. By comparing the spectra of the periods with and without type-B QPOs, we find that the spectral ratios above 10 keV are nearly constant or slightly decreasing, and the values are different between sources. Below 10 keV, the flux change of the Comptonization component is inversely proportional to the flux change of the thermal component, suggesting that the appearance of type-B QPOs is associated with a redistribution of the accretion power between the disc and the Comptonizing emission region. The spectral ratios between the periods with type-B QPO and those with broadband noise are significantly different from that with type-B QPO and without type-B QPO, where the ratios (type-B QPO/broadband noise) show a maximum at around 4 keV and then decrease gradually towards high energies. Finally, we discuss the possible change of the geometry of the inner accretion flow and/or jet during the transitions.

astro-ph.HE↗