Search

The phonon Boltzmann transport equation (BTE) is widely used for describing multiscale heat conduction (from nm to $\mu$m or mm) in solid materials. Developing numerical approaches to solve this equation is challenging since it is a 7-dimensional integral-differential equation. In this work, we propose Monte Carlo physics-informed neural networks (MC-PINNs), which do not suffer from the "curse of dimensionality", to solve the phonon BTE to model the multiscale heat conduction in solid materials. MC-PINNs use a deep neural network to approximate the solution to the BTE, and encode the BTE as well as the corresponding boundary/initial conditions using the automatic differentiation. In addition, we propose a novel two-step sampling approach to address inefficiency and inaccuracy issues in the widely used sampling methods in PINNs. In particular, we first randomly sample a certain number of points in the temporal-spatial space (Step I), and then draw another number of points randomly in the solid angular space (Step II). The training points are constructed using the tensor product of the data drawn from the above two steps. The two-step sampling strategy enables MC-PINNs (1) to model the heat conduction from ballistic to diffusive regimes, and (2) is more memory-efficient compared to conventional numerical solvers or existing PINNs for BTE. A series of numerical examples including quasi-one-dimensional (quasi-1D) steady/unsteady heat conduction in a film, and the heat conduction in a quasi-two- and three-dimensional square domain, are conducted to justify the effectiveness of the MC-PINN for heat conduction spanning diffusive and ballistic regimes. Finally, we compare the computational time and memory usage of the MC-PINN and a state-of-art numerical methods to demonstrate the potential of the MC-PINN for large scale problems in real-world applications.

Understanding the phonon transport mechanisms and efficiently capturing the spatiotemporal distributions of temperature is of great significance for alleviating hotspot issues in the electronic devices. Most previous simulations mainly focused on the steady-state problem with continuous heating, and the effective Fourier's law (EFL) is widely used for practical multiscale thermal engineering due to its simplicity and efficiency although it still follows the diffusive rule. However, non-continuous heating is more common in the electronic devices, and few comparative study is conducted to estimate how much deviation the EFL would produce. To answer above questions, the heat conduction in nanoscale bulk or silicon-on-insulator (SOI) transistors is investigated by the phonon Boltzmann transport equation (BTE) under three heating strategies, namely, `Continuous', `Intermittent' and `Alternating' heating. Numerical results in the quasi-2D or 3D hotspot systems show that it is not easy to accurately capture the micro/nano scale heat conduction by the EFL, especially near the hotspot regions. Different heating strategies have great influence on the temperature rise and transient thermal dissipation process. Compared to `Intermittent' heating, the temperature variance of `Alternating' heating is smaller., Comment: 25 pages, 52 figures

Due to flexibility and low-cost, unmanned aerial vehicles (UAVs) are increasingly crucial for enhancing coverage and functionality of wireless networks. However, incorporating UAVs into next-generation wireless communication systems poses significant challenges, particularly in sustaining high-rate and long-range secure communications against eavesdropping attacks. In this work, we consider a UAV swarm-enabled secure surveillance network system, where a UAV swarm forms a virtual antenna array to transmit sensitive surveillance data to a remote base station (RBS) via collaborative beamforming (CB) so as to resist mobile eavesdroppers. Specifically, we formulate an aerial secure communication and energy efficiency multi-objective optimization problem (ASCEE-MOP) to maximize the secrecy rate of the system and to minimize the flight energy consumption of the UAV swarm. To address the non-convex, NP-hard and dynamic ASCEE-MOP, we propose a generative diffusion model-enabled twin delayed deep deterministic policy gradient (GDMTD3) method. Specifically, GDMTD3 leverages an innovative application of diffusion models to determine optimal excitation current weights and position decisions of UAVs. The diffusion models can better capture the complex dynamics and the trade-off of the ASCEE-MOP, thereby yielding promising solutions. Simulation results highlight the superior performance of the proposed approach compared with traditional deployment strategies and some other deep reinforcement learning (DRL) benchmarks. Moreover, performance analysis under various parameter settings of GDMTD3 and different numbers of UAVs verifies the robustness of the proposed approach., Comment: This paper has been submitted to IEEE Transactions on Mobile Computing

Marine fog poses a significant hazard to global shipping, necessitating effective detection and forecasting to reduce economic losses. In recent years, several machine learning (ML) methods have demonstrated superior detection accuracy compared to traditional meteorological methods. However, most of these works are developed on proprietary datasets, and the few publicly accessible datasets are often limited to simplistic toy scenarios for research purposes. To advance the field, we have collected nearly a decade's worth of multi-modal data related to continuous marine fog stages from four series of geostationary meteorological satellites, along with meteorological observations and numerical analysis, covering 15 marine regions globally where maritime fog frequently occurs. Through pixel-level manual annotation by meteorological experts, we present the most comprehensive marine fog detection and forecasting dataset to date, named M4Fog, to bridge ocean and atmosphere. The dataset comprises 68,000 "super data cubes" along four dimensions: elements, latitude, longitude and time, with a temporal resolution of half an hour and a spatial resolution of 1 kilometer. Considering practical applications, we have defined and explored three meaningful tracks with multi-metric evaluation systems: static or dynamic marine fog detection, and spatio-temporal forecasting for cloud images. Extensive benchmarking and experiments demonstrate the rationality and effectiveness of the construction concept for proposed M4Fog. The data and codes are available to whole researchers through cloud platforms to develop ML-driven marine fog solutions and mitigate adverse impacts on human activities.

Accurately assessing the impact of electron-phonon interaction (EPI) on the lattice thermal conductivity of semiconductors is crucial for the thermal management of electronic devices and a unified physical understanding of this issue is highly desired. In this work, we predict the lattice thermal conductivities of typical direct and indirect bandgap semiconductors accounting for EPI based on mode-level first-principles calculations. It is found that EPI has a larger effect on the lattice thermal conductivity of $p$-type doping compared to $n$-type doping in the same semiconductor at high charge carrier concentrations. The stronger EPI in $p$-type doping is attributed to the relatively higher electron density of states caused by the relatively larger $p$-orbital component. Furthermore, EPI has a stronger influence on the lattice thermal conductivity of $n$-type indirect bandgap semiconductors than $n$-type direct bandgap semiconductors. This is attributed to the relatively lower electron density of states in direct bandgap semiconductors stemming from the $s$-orbital component. This work reveals that there exist diverse responses in lattice thermal conductivity of $n$-type/$p$-type semiconductors, which can be attributed to asymmetric EPIs., Comment: 8 pages,5 figures

4H-Silicon Carbide (4H-SiC) possesses a high Baliga figure of merit, making it a promising material for power electronics. However, its applications are limited by its low hole mobility. Herein, we found that the hole mobility of 4H-SiC is mainly limited by the strong interband electron-phonon scattering using mode-level first-principles calculations. Our research indicates that applying compressive strain can reverse the sign of crystal-field splitting and change the ordering of electron bands close to the valence band maximum. Therefore, the interband electron-phonon scattering is severely suppressed, and the out-of-plane hole mobility of 4H-SiC can be enhanced by 200% with 2% uniaxial compressive strain applied. This work provides new insights into the electron transport mechanisms in semiconductors and suggests a strategy to improve hole mobility that could be applied to other semiconductors with hexagonal crystalline geometries., Comment: 22 pages, 4 figures

Multimodal Entity Linking (MEL) aims to link ambiguous mentions in multimodal contexts to entities in a multimodal knowledge graph. A pivotal challenge is to fully leverage multi-element correlations between mentions and entities to bridge modality gap and enable fine-grained semantic matching. Existing methods attempt several local correlative mechanisms, relying heavily on the automatically learned attention weights, which may over-concentrate on partial correlations. To mitigate this issue, we formulate the correlation assignment problem as an optimal transport (OT) problem, and propose a novel MEL framework, namely OT-MEL, with OT-guided correlation assignment. Thereby, we exploit the correlation between multimodal features to enhance multimodal fusion, and the correlation between mentions and entities to enhance fine-grained matching. To accelerate model prediction, we further leverage knowledge distillation to transfer OT assignment knowledge to attention mechanism. Experimental results show that our model significantly outperforms previous state-of-the-art baselines and confirm the effectiveness of the OT-guided correlation assignment., Comment: Findings of ACL 2024

The well-established modular autonomous driving system is decoupled into different standalone tasks, e.g. perception, prediction and planning, suffering from information loss and error accumulation across modules. In contrast, end-to-end paradigms unify multi-tasks into a fully differentiable framework, allowing for optimization in a planning-oriented spirit. Despite the great potential of end-to-end paradigms, both the performance and efficiency of existing methods are not satisfactory, particularly in terms of planning safety. We attribute this to the computationally expensive BEV (bird's eye view) features and the straightforward design for prediction and planning. To this end, we explore the sparse representation and review the task design for end-to-end autonomous driving, proposing a new paradigm named SparseDrive. Concretely, SparseDrive consists of a symmetric sparse perception module and a parallel motion planner. The sparse perception module unifies detection, tracking and online mapping with a symmetric model architecture, learning a fully sparse representation of the driving scene. For motion prediction and planning, we review the great similarity between these two tasks, leading to a parallel design for motion planner. Based on this parallel design, which models planning as a multi-modal problem, we propose a hierarchical planning selection strategy , which incorporates a collision-aware rescore module, to select a rational and safe trajectory as the final planning output. With such effective designs, SparseDrive surpasses previous state-of-the-arts by a large margin in performance of all tasks, while achieving much higher training and inference efficiency. Code will be avaliable at https://github.com/swc-17/SparseDrive for facilitating future research.

Multi-scale learning is central to semantic segmentation. We visualize the effective receptive field (ERF) of canonical multi-scale representations and point out two risks in learning them: scale inadequacy and field inactivation. A novel multi-scale learner, varying window attention (VWA), is presented to address these issues. VWA leverages the local window attention (LWA) and disentangles LWA into the query window and context window, allowing the context's scale to vary for the query to learn representations at multiple scales. However, varying the context to large-scale windows (enlarging ratio R) can significantly increase the memory footprint and computation cost (R^2 times larger than LWA). We propose a simple but professional re-scaling strategy to zero the extra induced cost without compromising performance. Consequently, VWA uses the same cost as LWA to overcome the receptive limitation of the local window. Furthermore, depending on VWA and employing various MLPs, we introduce a multi-scale decoder (MSD), VWFormer, to improve multi-scale representations for semantic segmentation. VWFormer achieves efficiency competitive with the most compute-friendly MSDs, like FPN and MLP decoder, but performs much better than any MSDs. For instance, using nearly half of UPerNet's computation, VWFormer outperforms it by 1.0%-2.5% mIoU on ADE20K. With little extra overhead, ~10G FLOPs, Mask2Former armed with VWFormer improves by 1.0%-1.3%. The code and models are available at https://github.com/yan-hao-tian/vw, Comment: ICLR2024 Poster

Automatic detection of multimodal misinformation has gained a widespread attention recently. However, the potential of powerful Large Language Models (LLMs) for multimodal misinformation detection remains underexplored. Besides, how to teach LLMs to interpret multimodal misinformation in cost-effective and accessible way is still an open question. To address that, we propose MMIDR, a framework designed to teach LLMs in providing fluent and high-quality textual explanations for their decision-making process of multimodal misinformation. To convert multimodal misinformation into an appropriate instruction-following format, we present a data augmentation perspective and pipeline. This pipeline consists of a visual information processing module and an evidence retrieval module. Subsequently, we prompt the proprietary LLMs with processed contents to extract rationales for interpreting the authenticity of multimodal misinformation. Furthermore, we design an efficient knowledge distillation approach to distill the capability of proprietary LLMs in explaining multimodal misinformation into open-source LLMs. To explore several research questions regarding the performance of LLMs in multimodal misinformation detection tasks, we construct an instruction-following multimodal misinformation dataset and conduct comprehensive experiments. The experimental findings reveal that our MMIDR exhibits sufficient detection performance and possesses the capacity to provide compelling rationales to support its assessments., Comment: 10 pages, 3 figures

In recent years, temporal knowledge graph (TKG) reasoning has received significant attention. Most existing methods assume that all timestamps and corresponding graphs are available during training, which makes it difficult to predict future events. To address this issue, recent works learn to infer future events based on historical information. However, these methods do not comprehensively consider the latent patterns behind temporal changes, to pass historical information selectively, update representations appropriately and predict events accurately. In this paper, we propose the Historical Information Passing (HIP) network to predict future events. HIP network passes information from temporal, structural and repetitive perspectives, which are used to model the temporal evolution of events, the interactions of events at the same time step, and the known events respectively. In particular, our method considers the updating of relation representations and adopts three scoring functions corresponding to the above dimensions. Experimental results on five benchmark datasets show the superiority of HIP network, and the significant improvements on Hits@1 prove that our method can more accurately predict what is going to happen., Comment: 7 pages, 3 figures

Roadside perception can greatly increase the safety of autonomous vehicles by extending their perception ability beyond the visual range and addressing blind spots. However, current state-of-the-art vision-based roadside detection methods possess high accuracy on labeled scenes but have inferior performance on new scenes. This is because roadside cameras remain stationary after installation and can only collect data from a single scene, resulting in the algorithm overfitting these roadside backgrounds and camera poses. To address this issue, in this paper, we propose an innovative Scenario Generalization Framework for Vision-based Roadside 3D Object Detection, dubbed SGV3D. Specifically, we employ a Background-suppressed Module (BSM) to mitigate background overfitting in vision-centric pipelines by attenuating background features during the 2D to bird's-eye-view projection. Furthermore, by introducing the Semi-supervised Data Generation Pipeline (SSDG) using unlabeled images from new scenes, diverse instance foregrounds with varying camera poses are generated, addressing the risk of overfitting specific camera poses. We evaluate our method on two large-scale roadside benchmarks. Our method surpasses all previous methods by a significant margin in new scenes, including +42.57% for vehicle, +5.87% for pedestrian, and +14.89% for cyclist compared to BEVHeight on the DAIR-V2X-I heterologous benchmark. On the larger-scale Rope3D heterologous benchmark, we achieve notable gains of 14.48% for car and 12.41% for large vehicle. We aspire to contribute insights on the exploration of roadside perception techniques, emphasizing their capability for scenario generalization. The code will be available at https://github.com/yanglei18/SGV3D, Comment: 13 pages, 8 figures

Sparse meteorological forecasting is indispensable for fine-grained weather forecasting and deserves extensive attention. Recent studies have highlighted the potential of spatio-temporal graph convolutional networks (ST-GCNs) in predicting numerical data from ground weather stations. However, as one of the highest fidelity and lowest latency data, the application of the vision data from satellites in ST-GCNs remains unexplored. There are few studies to demonstrate the effectiveness of combining the above multi-modal data for sparse meteorological forecasting. Towards this objective, we introduce Vision-Numerical Fusion Graph Convolutional Network (VN-Net), which mainly utilizes: 1) Numerical-GCN (N-GCN) to adaptively model the static and dynamic patterns of spatio-temporal numerical data; 2) Vision-LSTM Network (V-LSTM) to capture multi-scale joint channel and spatial features from time series satellite images; 4) a GCN-based decoder to generate hourly predictions of specified meteorological factors. As far as we know, VN-Net is the first attempt to introduce GCN method to utilize multi-modal data for better handling sparse spatio-temporal meteorological forecasting. Our experiments on Weather2k dataset show VN-Net outperforms state-of-the-art by a significant margin on mean absolute error (MAE) and root mean square error (RMSE) for temperature, relative humidity, and visibility forecasting. Furthermore, we conduct interpretation analysis and design quantitative evaluation metrics to assess the impact of incorporating vision data.

A synthetic iterative scheme is developed for thermal applications in hotspot systems with large temperature variance. Different from previous work with linearized equilibrium state and small temperature difference assumption, the phonon equilibrium distribution shows a nonlinear relationship with temperature and mean free path changes with the spatial temperature when the temperature difference of system is large, so that the same phonon mode may suffer different transport processes in different geometric regions. In order to efficiently capture nonlinear and multiscale thermal behaviors, the Newton method is used and a macroscopic iteration is introduced for preprocessing based on the iterative solutions of the stationary phonon BTE. Macroscopic and mesoscopic physical evolution processes are connected by the heat flux, which is no longer calculated by classical Fourier's law but obtained by taking the moment of phonon distribution function. These two processes exchange information from different scales, such that the present scheme could efficiently deal with heat conduction problems from ballistic to diffusive regime. Numerical tests show that the present scheme could efficiently capture the multiscale heat conduction in hotspot systems with large temperature variances. In addition, a comparison is made between the solutions of the present scheme and effective Fourier's law by several heat dissipations problems under different sizes or selective phonon excitation. Numerical results show that compared to the classical Fourier's law, the results of the effective Fourier's law could be closer to the BTE solutions by adjusting effective coefficients. However, it is still difficult to capture some local nonlinear phenomena in complex geometries., Comment: 20 pages, 43 figures

Electrons are the carriers of heat and electricity in materials, and exhibit abundant transport phenomena such as ballistic, diffusive, and hydrodynamic behaviors in systems with different sizes. The electron Boltzmann transport equation (eBTE) is a reliable model for describing electron transport, but it is a challenging problem to efficiently obtain the numerical solutions of eBTE within one unified scheme involving ballistic, hydrodynamics and/or diffusive regimes. In this work, a discrete unified gas kinetic scheme (DUGKS) in finite-volume framework is developed based on the eBTE with the Callaway relaxation model for electron transport. By reconstructing the distribution function at the cell interface, the processes of electron drift and scattering are coupled together within a single time step. Numerical tests demonstrate that the DUGKS can be adaptively applied to multiscale electron transport, across different regimes.

Unmanned aerial vehicles (UAVs) as aerial relays are practically appealing for assisting Internet of Things (IoT) network. In this work, we aim to utilize the UAV swarm to assist the secure communication between the micro base station (MBS) equipped with the planar array antenna (PAA) and the IoT terminal devices by collaborative beamforming (CB), so as to counteract the effects of collusive eavesdropping attacks in time-domain. Specifically, we formulate a UAV swarm-enabled secure relay multi-objective optimization problem (US2RMOP) for simultaneously maximizing the achievable sum rate of associated IoT terminal devices, minimizing the achievable sum rate of the eavesdropper and minimizing the energy consumption of UAV swarm, by jointly optimizing the excitation current weights of both MBS and UAV swarm, the selection of the UAV receiver, the position of UAVs and user association order of IoT terminal devices. Furthermore, the formulated US2RMOP is proved to be a non-convex, NP-hard and large-scale optimization problem. Therefore, we propose an improved multi-objective grasshopper algorithm (IMOGOA) with some specific designs to address the problem. Simulation results exhibit the effectiveness of the proposed UAV swarm-enabled collaborative secure relay strategy and demonstrate the superiority of IMOGOA., Comment: Submitted to IEEE Transactions on Mobile Computing

Although the majority of recent autonomous driving systems concentrate on developing perception methods based on ego-vehicle sensors, there is an overlooked alternative approach that involves leveraging intelligent roadside cameras to help extend the ego-vehicle perception ability beyond the visual range. We discover that most existing monocular 3D object detectors rely on the ego-vehicle prior assumption that the optical axis of the camera is parallel to the ground. However, the roadside camera is installed on a pole with a pitched angle, which makes the existing methods not optimal for roadside scenes. In this paper, we introduce a novel framework for Roadside Monocular 3D object detection with ground-aware embeddings, named MonoGAE. Specifically, the ground plane is a stable and strong prior knowledge due to the fixed installation of cameras in roadside scenarios. In order to reduce the domain gap between the ground geometry information and high-dimensional image features, we employ a supervised training paradigm with a ground plane to predict high-dimensional ground-aware embeddings. These embeddings are subsequently integrated with image features through cross-attention mechanisms. Furthermore, to improve the detector's robustness to the divergences in cameras' installation poses, we replace the ground plane depth map with a novel pixel-level refined ground plane equation map. Our approach demonstrates a substantial performance advantage over all previous monocular 3D object detectors on widely recognized 3D detection benchmarks for roadside cameras. The code and pre-trained models will be released soon., Comment: 12 pages, 6 figures

The electron-phonon coupling in ultrafast heating systems is studied within the framework of Boltzmann transport equation (BTE) with coupled electron and phonon transport. A discrete unified gas kinetic scheme is developed to solve the BTE, in which the electron/phonon advection, scattering and electron-phonon interactions are coupled together within one time step by solving the BTE again at the cell interface. Numerical results show that the present scheme can correctly predict the electron-phonon coupling constant, and is in excellent agreement with typical two-temperature model (TTM) and experimental results in existing literatures and our performed time-domain thermoreflectance technique. It can also capture the ballistic or thermal wave effects when the characteristic length/time is comparable to or smaller than the mean free path/relaxation time where the TTM fails. Finally, the electron-phonon coupling in transient thermal grating geometry and Au/Pt bilayer metals with interfacial thermal resistance is simulated and discussed. For the former, heat flow from phonon to electron is predicted in both the ballistic and diffusive regimes. For the latter, the reflected signal increases in the early tens of picoseconds and then decreases with time after the heat source is removed., Comment: 16 pages, 12 figures

Automatically extracting roads from satellite imagery is a fundamental yet challenging computer vision task in the field of remote sensing. Pixel-wise semantic segmentation-based approaches and graph-based approaches are two prevailing schemes. However, prior works show the imperfections that semantic segmentation-based approaches yield road graphs with low connectivity, while graph-based methods with iterative exploring paradigms and smaller receptive fields focus more on local information and are also time-consuming. In this paper, we propose a new scheme for multi-task satellite imagery road extraction, Patch-wise Road Keypoints Detection (PaRK-Detect). Building on top of D-LinkNet architecture and adopting the structure of keypoint detection, our framework predicts the position of patch-wise road keypoints and the adjacent relationships between them to construct road graphs in a single pass. Meanwhile, the multi-task framework also performs pixel-wise semantic segmentation and generates road segmentation masks. We evaluate our approach against the existing state-of-the-art methods on DeepGlobe, Massachusetts Roads, and RoadTracer datasets and achieve competitive or better results. We also demonstrate a considerable outperformance in terms of inference speed., Comment: Accepted at BMVC 2022 (Oral). 13 pages, 5 figures. https://bmvc2022.mpi-inf.mpg.de/381/

Automatic detection of multimodal fake news has gained a widespread attention recently. Many existing approaches seek to fuse unimodal features to produce multimodal news representations. However, the potential of powerful cross-modal contrastive learning methods for fake news detection has not been well exploited. Besides, how to aggregate features from different modalities to boost the performance of the decision-making process is still an open question. To address that, we propose COOLANT, a cross-modal contrastive learning framework for multimodal fake news detection, aiming to achieve more accurate image-text alignment. To further improve the alignment precision, we leverage an auxiliary task to soften the loss term of negative samples during the contrast process. A cross-modal fusion module is developed to learn the cross-modality correlations. An attention mechanism with an attention guidance module is implemented to help effectively and interpretably aggregate the aligned unimodal representations and the cross-modality correlations. Finally, we evaluate the COOLANT and conduct a comparative study on two widely used datasets, Twitter and Weibo. The experimental results demonstrate that our COOLANT outperforms previous approaches by a large margin and achieves new state-of-the-art results on the two datasets., Comment: 9 pages, 3 figures

The creation of a parameterized stylized character involves careful selection of numerous parameters, also known as the "avatar vectors" that can be interpreted by the avatar engine. Existing unsupervised avatar vector estimation methods that auto-create avatars for users, however, often fail to work because of the domain gap between realistic faces and stylized avatar images. To this end, we propose SwiftAvatar, a novel avatar auto-creation framework that is evidently superior to previous works. SwiftAvatar introduces dual-domain generators to create pairs of realistic faces and avatar images using shared latent codes. The latent codes can then be bridged with the avatar vectors as pairs, by performing GAN inversion on the avatar images rendered from the engine using avatar vectors. Through this way, we are able to synthesize paired data in high-quality as many as possible, consisting of avatar vectors and their corresponding realistic faces. We also propose semantic augmentation to improve the diversity of synthesis. Finally, a light-weight avatar vector estimator is trained on the synthetic pairs to implement efficient auto-creation. Our experiments demonstrate the effectiveness and efficiency of SwiftAvatar on two different avatar engines. The superiority and advantageous flexibility of SwiftAvatar are also verified in both subjective and objective evaluations., Comment: AAAI 2023 Oral

Many macroscopic non-Fourier heat conduction models have been developed in the past decades based on Chapman-Enskog, Hermite or other small perturbation expansion methods. These macroscopic models have made great success on capturing non-Fourier thermal behaviors in solid materials, but most of them are limited by small Knudsen numbers and incapable of capturing highly non-equilibrium or ballistic thermal transport. In this paper, we provide a new strategy for constructing macroscopic non-Fourier heat conduction modeling, that is, using data-driven deep learning methods combined with non-equilibrium thermodynamics instead of small perturbation expansion. We present the mechanism-data fusion method, an approach that seamlessly integrates the rigorous framework of Conservation-Dissipation Formalism (CDF) with the flexibility of machine learning to model non-Fourier heat conduction. Leveraging the conservation-dissipation principle with dual-dissipative variables, we derive an interpretable series of partial differential equations, fine-tuned through a training strategy informed by data from the phonon Boltzmann transport equation. Moreover, we also present the inner-step operation to narrow the gap from the discrete form to the continuous system. Through numerical tests, our model demonstrates excellent predictive capabilities across various heat conduction regimes, including diffusive, hydrodynamic and ballistic regimes, and displays its robustness and precision even with discontinuous initial conditions.

Mesoscopic numerical simulation has become an important tool in thermal management and energy harvesting at the micro/nano scale, where the Fourier's law failed. However, it is not easy to efficiently solve the phonon Boltzmann transport equation (BTE) from ballistic to diffusive limit. In order to accelerate convergence, an implicit synthetic iterative scheme is developed for the stationary phonon BTE, in which a macroscopic moment equation is invoked and solved iteratively coupled with the typical source iteration of the kinetic equation. Different from previous numerical interpolation, the phonon BTE is solved again at the cell interface along the group velocity direction within a certain length when reconstructing the interfacial phonon distribution function. Fourier stability analysis shows that the present method could converge faster than the source iteration method in the (near) diffusive regime. Numerical results prove that the present scheme can capture the ballistic-diffusive effects correctly and efficiently. The present acceleration framework could be a powerful tool for simulating practical thermal engineering problems in the future., Comment: 25 pages, 19 figures, 54 references

Performance of trimap-free image matting methods is limited when trying to decouple the deterministic and undetermined regions, especially in the scenes where foregrounds are semantically ambiguous, chromaless, or high transmittance. In this paper, we propose a novel framework named Privileged Prior Information Distillation for Image Matting (PPID-IM) that can effectively transfer privileged prior environment-aware information to improve the performance of students in solving hard foregrounds. The prior information of trimap regulates only the teacher model during the training stage, while not being fed into the student network during actual inference. In order to achieve effective privileged cross-modality (i.e. trimap and RGB) information distillation, we introduce a Cross-Level Semantic Distillation (CLSD) module that reinforces the trimap-free students with more knowledgeable semantic representations and environment-aware information. We also propose an Attention-Guided Local Distillation module that efficiently transfers privileged local attributes from the trimap-based teacher to trimap-free students for the guidance of local-region optimization. Extensive experiments demonstrate the effectiveness and superiority of our PPID framework on the task of image matting. In addition, our trimap-free IndexNet-PPID surpasses the other competing state-of-the-art methods by a large margin, especially in scenarios with chromaless, weak texture, or irregular objects., Comment: 15 pages, 7 figures

In this work, a steady discrete unified gas kinetic scheme (SDUGKS) is proposed to solve the steady radiative transfer equation (RTE), which is an improvement of the original SDUGKS [X. F. Zhou et al., J. Comput. Phys. 423, 109767 (2020)]. The trapezoidal rule other than the rectangular rule used in the original SDUGKS is adopted in the proposed method in the reconstruction of energy flux across cell interface, just as the unsteady DUGKS. By this way, the characteristic line length of the modified SDUGKS establishes a relationship with the Courant-Friedrichs-Lewy (CFL) number in the DUGKS, which guarantees the accuracy of the modified SDUGKS. Furthermore, the characteristic line length is no longer limited by the extinction coefficient like in original SDUGKS. As a result, the modified SDUGKS is more accurate and robust than original SDUGKS, and more efficient than the DUGKS for steady radiation problems. Furthermore, the smooth linear interpolation and the van Leer limiter are used for problems with smooth and discontinuous optical thicknesses, respectively. Several numerical tests with optical thickness varying from optical thin to thick are conducted to validate the present scheme. Numerical results demonstrate that the modified SDUGKS can serve as an effective tool in the study of multiscale steady radiative heat transfer in participating media., Comment: 23pages,16 figures,2 tables

Monocular 3D object detection is an essential perception task for autonomous driving. However, the high reliance on large-scale labeled data make it costly and time-consuming during model optimization. To reduce such over-reliance on human annotations, we propose Mix-Teaching, an effective semi-supervised learning framework applicable to employ both labeled and unlabeled images in training stage. Mix-Teaching first generates pseudo-labels for unlabeled images by self-training. The student model is then trained on the mixed images possessing much more intensive and precise labeling by merging instance-level image patches into empty backgrounds or labeled images. This is the first to break the image-level limitation and put high-quality pseudo labels from multi frames into one image for semi-supervised training. Besides, as a result of the misalignment between confidence score and localization quality, it's hard to discriminate high-quality pseudo-labels from noisy predictions using only confidence-based criterion. To that end, we further introduce an uncertainty-based filter to help select reliable pseudo boxes for the above mixing operation. To the best of our knowledge, this is the first unified SSL framework for monocular 3D object detection. Mix-Teaching consistently improves MonoFlex and GUPNet by significant margins under various labeling ratios on KITTI dataset. For example, our method achieves around +6.34% AP@0.7 improvement against the GUPNet baseline on validation set when using only 10% labeled data. Besides, by leveraging full training set and the additional 48K raw images of KITTI, it can further improve the MonoFlex by +4.65% improvement on AP@0.7 for car detection, reaching 18.54% AP@0.7, which ranks the 1st place among all monocular based methods on KITTI test leaderboard. The code and pretrained models will be released at https://github.com/yanglei18/Mix-Teaching., Comment: 11 pages, 5 figures, 7 tables