Search

This work presents a LiDAR-based quadrotor system for slope inspection in dense vegetation environments. Cities like Hong Kong are vulnerable to climate hazards, which often result in landslides. To mitigate the landslide risks, the Civil Engineering and Development Department (CEDD) has constructed steel flexible debris-resisting barriers on vulnerable natural catchments to protect residents. However, it is necessary to carry out regular inspections to identify any anomalies, which may affect the proper functioning of the barriers. Traditional manual inspection methods face challenges and high costs due to steep terrain and dense vegetation. Compared to manual inspection, unmanned aerial vehicles (UAVs) equipped with LiDAR sensors and cameras have advantages such as maneuverability in complex terrain, and access to narrow areas and high spots. However, conducting slope inspections using UAVs in dense vegetation poses significant challenges. First, in terms of hardware, the overall design of the UAV must carefully consider its maneuverability in narrow spaces, flight time, and the types of onboard sensors required for effective inspection. Second, regarding software, navigation algorithms need to be designed to enable obstacle avoidance flight in dense vegetation environments. To overcome these challenges, we develop a LiDAR-based quadrotor, accompanied by a comprehensive software system. The goal is to deploy our quadrotor in field environments to achieve efficient slope inspection. To assess the feasibility of our hardware and software system, we conduct functional tests in non-operational scenarios. Subsequently, invited by CEDD, we deploy our quadrotor in six field environments, including five flexible debris-resisting barriers located in dense vegetation and one slope that experienced a landslide. These experiments demonstrated the superiority of our quadrotor in slope inspection., Comment: 36 pages

In this paper, the inverse scattering transform for the integrable discrete nonlocal PT symmetric nonlinear Schr\"odinger equation with nonzero boundary conditions is presented. According to the two different signs of symmetry reduction and two different values of the phase difference between plus and minus infinity, we discuss four cases with significant differences about analytical regions, symmetry, asymptotic behavior and the presence or absence of discrete eigenvalues, namely, the existence or absence of soliton solutions. Therefore, in all cases, we study the direct scattering and inverse scattering problem, separately. The Riemann-Hilbert problem is constructed and solved as well as the reconstruction formula of potential is derived, respectively. Finally, combining the time evolution, we provide the dark, bright, dark-bright soliton solutions on the nonzero background in difference cases under the reflectionless condition.

In this paper, the nonlocal reverse space-time derivative nonlinear Schr\"odinger equation under nonzero boundary conditions is investigated using the Riemann-Hilbert (RH) approach. The direct scattering problem focuses on the analyticity, symmetries, and asymptotic behaviors of the Jost eigenfunctions and scattering matrix functions, leading to the construction of the corresponding RH problem. Then, in the inverse scattering problem, the Plemelj formula is employed to solve the RH problem. So the reconstruction formula, trace formulae, $\theta$ condition, and exact expression of the single-pole and double-pole solutions are obtained. Furthermore, dark-dark solitons, bright-dark solitons, and breather solutions of the reverse space-time derivative nonlinear Schr\"odinger equation are presented along with their dynamic behaviors summarized through graphical simulation.

The BioNLP ACL'24 Shared Task on Streamlining Discharge Documentation aims to reduce the administrative burden on clinicians by automating the creation of critical sections of patient discharge letters. This paper presents our approach using the Llama3 8B quantized model to generate the "Brief Hospital Course" and "Discharge Instructions" sections. We employ a zero-shot method combined with Retrieval-Augmented Generation (RAG) to produce concise, contextually accurate summaries. Our contributions include the development of a curated template-based approach to ensure reliability and consistency, as well as the integration of RAG for word count prediction. We also describe several unsuccessful experiments to provide insights into our pathway for the competition. Our results demonstrate the effectiveness and efficiency of our approach, achieving high scores across multiple evaluation metrics., Comment: BioNLP 2024 workshop

In cardiac Magnetic Resonance Imaging (MRI) analysis, simultaneous myocardial segmentation and T2 quantification are crucial for assessing myocardial pathologies. Existing methods often address these tasks separately, limiting their synergistic potential. To address this, we propose SQNet, a dual-task network integrating Transformer and Convolutional Neural Network (CNN) components. SQNet features a T2-refine fusion decoder for quantitative analysis, leveraging global features from the Transformer, and a segmentation decoder with multiple local region supervision for enhanced accuracy. A tight coupling module aligns and fuses CNN and Transformer branch features, enabling SQNet to focus on myocardium regions. Evaluation on healthy controls (HC) and acute myocardial infarction patients (AMI) demonstrates superior segmentation dice scores (89.3/89.2) compared to state-of-the-art methods (87.7/87.9). T2 quantification yields strong linear correlations (Pearson coefficients: 0.84/0.93) with label values for HC/AMI, indicating accurate mapping. Radiologist evaluations confirm SQNet's superior image quality scores (4.60/4.58 for segmentation, 4.32/4.42 for T2 quantification) over state-of-the-art methods (4.50/4.44 for segmentation, 3.59/4.37 for T2 quantification). SQNet thus offers accurate simultaneous segmentation and quantification, enhancing cardiac disease diagnosis, such as AMI., Comment: 10 pages, 8 figures, 6 tables

Numerical simulations indicate that correlations exist between the velocity distributions of stars and dark matter (DM). We study the local DM velocity distribution based on these correlations. We select K giants from LAMOST DR8 cross-matched with Gaia DR3, which has robust measurements of three-dimensional velocity and metallicity, and separate them into the disk, halo substructure and main halo components in the chemo-dynamical space utilizing the Gaussian Mixture Model. The substructure component is highly radially anisotropic, and possibly related to the Gaia-Enceladus-Sausage (GES) merger event, while the halo component is isotropic and accreted from the earliest mergers following the Maxwell-Boltzmann Distribution (Standard Halo Model, SHM). We find that the GES-like substructure contributes $\sim85\%$ of the local non-disk stars in the Solar neighbourhood, which is nearly invariant when applying different volume cuts or additional angular momentum constraints. Utilizing the metallicity-stellar-mass relation and the stellar-mass-halo-mass relation, we find that $\sim25_{-15}^{+24}\%$ of local DM is in the kinematic substructure. Combined with the stellar distributions of non-disk components, we compute the velocity distribution of local DM. The modified heliocentric velocity distribution of local DM shifts to a lower speed and has a sharper peak compared to the SHM, which yields updated detection limits for the DM direct detection experiments. Our work confirms that the local DM velocity distribution deviates from the SHM, and needs to be properly accounted in the DM detection experiments.

Quiescent galaxies generally possess denser cores than star-forming galaxies with similar mass. As a measurement of the core density, the central stellar mass surface density within a radius of 1 kpc ($\Sigma_1$) was thus suggested to be closely related to galaxy quenching. Massive star-forming galaxies with high $\Sigma_1$ do not fit into this picture. To understand the origin of such galaxies, we compare the spatially-resolved stellar population and star formation properties of massive ($ > 10^{10.5}{\rm M}_{\odot}$) blue spiral galaxies with high and low $\Sigma_1$, divided by $\Sigma_1 = 10^{9.4} M_\odot \, {\rm kpc}^{-2}$, based on the final release of MaNGA IFU data. We find that both high $\Sigma_1$ and low $\Sigma_1$ blue spirals show large diversities in stellar population and star formation properties. Despite the diversities, high $\Sigma_1$ blue spirals are statistically different from the low $\Sigma_1$ ones. Specifically, the radial profiles of the luminosity-weighted age and Mgb/${\rm \langle Fe \rangle}$ show that high $\Sigma_1$ blue spirals consist of a larger fraction of galaxies with younger and less $\alpha$-element enhanced centers than their low $\Sigma_1$ counterparts, $\sim 55\%$ versus $\sim 30\%$. The galaxies with younger centers mostly have higher central specific star formation rates, which still follow the spaxel-based star formation main sequence relation though. Examinations of the H$\alpha$ velocity field and the optical structures suggest that galactic bars or galaxy interactions should be responsible for the rejuvenation of these galaxies. The remaining $\sim 45\% $ of high $\Sigma_1$ blue spirals are consistent with the inside-out growth scenario., Comment: 22 pages, 14 figures, accepted for publication in ApJ

Under investigation in this paper is the fractional integrable and non-integrable discrete modified Korteweg-de Vries hierarchies. The linear dispersion relations, completeness relations, inverse scattering transform, and fractional soliton solutions of the fractional integrable discrete modified Korteweg-de Vries hierarchy will be explored. The inverse scattering problem will be solved accurately by using Gel'fand-Levitan-Marchenko (GLM) equations and Riemann-Hilbert (RH) problem. The peak velocity of fractional soliton solutions will be analyzed. The numerical solutions of the non-integrable fractional averaged discrete modified Korteweg-de Vries equation which has a simpler form than the integrable one will be obtained by a split-step fourier method.

We focus on the semi-discrete complex modified Korteweg-de Vries (DcmKdV) equation in this paper. The direct and inverse scattering theory is developed with zero and non-zero boundary conditions (BCs) of the potential. For direct problem, the properties of the eigenfunctions and the scattering matrix, including analyticity, asymptotics and symmetries, are investigated, which facilitates the establishment of the Riemann-Hilbert (RH) problems. By solving the RH problems in the inverse problem part, the reconstruction potential formulas are obtained, which allows us to derive the N-soliton solutions in the reflectionless case. Meanwhile, the trace formulas are derived by means of studying the corresponding RH problems. Furthermore, the dynamic characteristics of the 1-soliton and 2-soliton with zero and non-zero boundary are demonstrated by graphical simulation.

Serving systems for Large Language Models (LLMs) are often optimized to improve quality of service (QoS) and throughput. However, due to the lack of open-sourced LLM serving workloads, these systems are frequently evaluated under unrealistic workload assumptions. Consequently, performance may degrade when these systems are deployed in real-world scenarios. This work presents BurstGPT, an LLM serving workload with 5.29 million traces from regional Azure OpenAI GPT services over 121 days. BurstGPT captures realistic LLM serving characteristics through detailed tracing of: (1) Concurrency of requests: It traces burstiness variations of requests in Azure OpenAI GPT services, revealing diversified concurrency patterns in different services and model types. (2) Response Lengths of requests: It traces the auto-regressive serving processes of GPT models, showing statistical relations between requests and their responses. (3) Failures of requests: It traces failures of conversation and API services, showing intensive resource needs and limited resource availability of such services in Azure. Details of the characteristics can serve multiple purposes in LLM serving optimizations, such as system evaluation and trace provisioning. In our demo evaluation with BurstGPT, we observe that frequent variations in BurstGPT reveal declines in efficiency, stability, or reliability in realistic LLM serving. We identify that the generalization of KV cache management and request scheduling optimization is not guaranteed for different workloads, especially when systems are poorly optimized for unrealistic workloads. We have made the dataset publicly available to encourage further research at https://github.com/HPMLL/BurstGPT.

We select 1,052,469 (754,635) thin disk stars from {\it Gaia} eDR3 and LAMOST DR7 in the range of Galactocentric radius $R$ (guiding center radius $R_\mathrm{g}$) from 8 to 11\,kpc to investigate the asymmetries between the North and South of the disk midplane. More specifically we analyze the vertical velocity dispersion profiles ($\sigma_{v_{z}}(z$)) in different bins of $R$ ($R_\mathrm{g}$) and $[\mathrm{Fe/H}]$. We find troughs in the profiles of $\sigma_{v_{z}}(z)$ located in both the North ($z \sim 0.7$\,kpc) and South ($z \sim -0.5$\,kpc) of the disk at all radial and chemical bins studied. The difference between the Northern and Southern vertical velocity dispersion profiles ($\Delta\sigma_{v_{z}}(|z|)$) shows a shift between curves of different $R$ and $R_\mathrm{g}$. A similar shift exists in these NS asymmetry profiles further divided into different $[\mathrm{Fe/H}]$ ranges. The sample binned with $R_\mathrm{g}$ more clearly displays the features in the velocity dispersion profiles. The shift in the peaks of the $\Delta\sigma_{v_{z}}$ profiles and the variation in the phase spiral shape binned by metallicity indicate the variation of the vertical potential profiles and the radial metallicity gradient. The wave-like signal in NS asymmetry of $\sigma_{v_{z}}(z)$ largely originates from phase spiral; while the NS asymmetry profiles of [Fe/H] only display a weak wave-like feature near solar radius. We perform a test particle simulation to qualitatively reproduce the observed results. A quantitative explanation of the NS asymmetry in the metallicity profile needs careful consideration of the spiral shape and the perturbation model, and we leave this for future work., Comment: 13 pages, 12 figures, accepted for publication in MNRAS