Search

In medical time series disease diagnosis, two key challenges are identified.First, the high annotation cost of medical data leads to overfitting in models trained on label-limited, single-center datasets. To address this, we propose incorporating external data from related tasks and leveraging AE-GAN to extract prior knowledge,providing valuable references for downstream tasks. Second, many existing studies employ contrastive learning to derive more generalized medical sequence representations for diagnostic tasks, usually relying on manually designed diverse positive and negative sample pairs.However, these approaches are complex, lack generalizability, and fail to adaptively capture disease-specific features across different conditions.To overcome this, we introduce LMCF (Learnable Multi-views Contrastive Framework), a framework that integrates a multi-head attention mechanism and adaptively learns representations from different views through inter-view and intra-view contrastive learning strategies.Additionally, the pre-trained AE-GAN is used to reconstruct discrepancies in the target data as disease probabilities, which are then integrated into the contrastive learning process.Experiments on three target datasets demonstrate that our method consistently outperforms seven other baselines, highlighting its significant impact on healthcare applications such as the diagnosis of myocardial infarction, Alzheimer's disease, and Parkinson's disease., Comment: 15 pages,6 figures

Limited by the operation and measurement conditions, numerical simulation is often the only feasible approach for studying plasma behavior and mechanisms. Although artificial intelligence methods, especially physics-informed neural network (PINN), have been widely applied in plasma simulation, the design of the neural network structures still largely relies on the experience of researchers. Meanwhile, existing neural architecture search methods tailored for PINN have encountered failures when dealing with complex plasma governing equations characterized by variable coefficients and strong nonlinearity. Therefore, we propose an improved neural architecture search-guided method, namely NAS-PINNv2, to address the limitations of existing methods. By analyzing the causes of failure, the sigmoid function is applied to calculate the architecture-related weights, and a new loss term is introduced. The performance of NAS-PINNv2 is verified in several numerical experiments including the Elenbaas-Heller equation without and with radial velocity, the drift-diffusion-Poisson equation and the Boltzmann equation. The results again emphasize that larger neural networks do not necessarily perform better, and the discovered neural architecture with multiple neuron numbers in a single hidden layer imply a more flexible and sophisticated design rule for fully connected networks., Comment: 15 pages, 10 figures, 4 tables

Hypergraph neural networks have been promising tools for handling learning tasks involving higher-order data, with notable applications in web graphs, such as modeling multi-way hyperlink structures and complex user interactions. Yet, their generalization abilities in theory are less clear to us. In this paper, we seek to develop margin-based generalization bounds for four representative classes of hypergraph neural networks, including convolutional-based methods (UniGCN), set-based aggregation (AllDeepSets), invariant and equivariant transformations (M-IGN), and tensor-based approaches (T-MPHN). Through the PAC-Bayes framework, our results reveal the manner in which hypergraph structure and spectral norms of the learned weights can affect the generalization bounds, where the key technical challenge lies in developing new perturbation analysis for hypergraph neural networks, which offers a rigorous understanding of how variations in the model's weights and hypergraph structure impact its generalization behavior. Our empirical study examines the relationship between the practical performance and theoretical bounds of the models over synthetic and real-world datasets. One of our primary observations is the strong correlation between the theoretical bounds and empirical loss, with statistically significant consistency in most cases., Comment: The Web Conference 2025

The essence of audio-visual segmentation (AVS) lies in locating and delineating sound-emitting objects within a video stream. While Transformer-based methods have shown promise, their handling of long-range dependencies struggles due to quadratic computational costs, presenting a bottleneck in complex scenarios. To overcome this limitation and facilitate complex multi-modal comprehension with linear complexity, we introduce AVS-Mamba, a selective state space model to address the AVS task. Our framework incorporates two key components for video understanding and cross-modal learning: Temporal Mamba Block for sequential video processing and Vision-to-Audio Fusion Block for advanced audio-vision integration. Building on this, we develop the Multi-scale Temporal Encoder, aimed at enhancing the learning of visual features across scales, facilitating the perception of intra- and inter-frame information. To perform multi-modal fusion, we propose the Modality Aggregation Decoder, leveraging the Vision-to-Audio Fusion Block to integrate visual features into audio features across both frame and temporal levels. Further, we adopt the Contextual Integration Pyramid to perform audio-to-vision spatial-temporal context collaboration. Through these innovative contributions, our approach achieves new state-of-the-art results on the AVSBench-object and AVSBench-semantic datasets. Our source code and model weights are available at AVS-Mamba., Comment: Accepted to IEEE Transactions on Multimedia (TMM)

Chain-of-Thought (CoT) reasoning is widely used to enhance the mathematical reasoning capabilities of large language models (LLMs). The introduction of process supervision for CoT trajectories has sparked discussions on improving test-time scaling, thereby unlocking the System 2-style thinking capabilities of these models. However, in multimodal mathematical reasoning, the scarcity of high-quality CoT training data has hindered existing models from achieving both deliberate reasoning and fine-grained verification. In this work, we propose a novel framework that introduces System 2-style thinking to multimodal mathematical reasoning. We introduce a three-module CoT data synthesis process that integrates CoT distillation, trajectory-format rewriting, and format unification. This process generates MMathCoT-1M, a high-quality CoT reasoning instruction fine-tuning dataset. Furthermore, we implement a dual-view trajectory labeling automation that targets both visual grounding fidelity and deductive chain validity, resulting in the DualMath-1.1M dataset. The URSA-8B model, trained on MMathCoT-1M, achieves new state-of-the-art (SOTA) performance among similarly sized multimodal LLMs on six popular reasoning benchmarks. Training URSA-8B further on the DualMath-1.1M dataset yields URSA-RM-8B, a verifier that enhances URSA-8B's test-time performance and surpasses strong closed-source multimodal MLLMs like GPT-4o. The model weights, training data, and code have been open-sourced: https://github.com/URSA-MATH/URSA-MATH., Comment: Add benchmarks and baselines. 27 pages, 11 tables, 17 figures. Models, training data and code have been open-sourced. Project url: https://ursa-math.github.io

Recommender systems are widely used in various real-world applications, but they often encounter the persistent challenge of the user cold-start problem. Cross-domain recommendation (CDR), which leverages user interactions from one domain to improve prediction performance in another, has emerged as a promising solution. However, users with similar preferences in the source domain may exhibit different interests in the target domain. Therefore, directly transferring embeddings may introduce irrelevant source-domain collaborative information. In this paper, we propose a novel graph-based disentangled contrastive learning framework to capture fine-grained user intent and filter out irrelevant collaborative information, thereby avoiding negative transfer. Specifically, for each domain, we use a multi-channel graph encoder to capture diverse user intents. We then construct the affinity graph in the embedding space and perform multi-step random walks to capture high-order user similarity relationships. Treating one domain as the target, we propose a disentangled intent-wise contrastive learning approach, guided by user similarity, to refine the bridging of user intents across domains. Extensive experiments on four benchmark CDR datasets demonstrate that DisCo consistently outperforms existing state-of-the-art baselines, thereby validating the effectiveness of both DisCo and its components., Comment: Accepted at AAAI 2025

In two dimensions, Hermitian lattices with non-zero Chern numbers and non-Hermitian lattices with a higher-order skin effect (HOSE) bypass the constraints of the Nielsen-Ninomiya "no-go" theorem at their one-dimensional boundaries. This allows the realization of topologically protected one-dimensional edge states with nonreciprocal dynamics. However, unlike the edge states of Chern insulators, the nonreciprocal edges of HOSE phases exist only at certain edges of the two-dimensional lattice, not all, leading to corner-localized states. In this work, we investigate the topological connections between these two systems and uncover novel non-Hermitian topological phases possessing "higher-order topological knots" (HOTKs). These phases arise from multiband topology protected by crystalline symmetries and host point-gap-protected nonreciprocal edge states that circulate around the entire boundary of the two-dimensional lattice. We show that phase transitions typically separate HOTK phases from "Complex Chern insulator" phases --non-Hermitian lattices with nonzero Chern numbers protected by imaginary line gaps in the presence of time-reversal symmetry., Comment: 22 pages, 17 figures

Channel estimation is a fundamental task in communication systems and is critical for effective demodulation. While most works deal with a simple scenario where the measurements are corrupted by the additive white Gaussian noise (AWGN), this work addresses the more challenging scenario where both AWGN and structured interference coexist. Such conditions arise, for example, when a sonar/radar transmitter and a communication receiver operate simultaneously within the same bandwidth. To ensure accurate channel estimation in these scenarios, the sparsity of the channel in the delay domain and the complicate structure of the interference are jointly exploited. Firstly, the score of the structured interference is learned via a neural network based on the diffusion model (DM), while the channel prior is modeled as a Gaussian distribution, with its variance controlling channel sparsity, similar to the setup of the sparse Bayesian learning (SBL). Then, two efficient posterior sampling methods are proposed to jointly estimate the sparse channel and the interference. Nuisance parameters, such as the variance of the prior are estimated via the expectation maximization (EM) algorithm. The proposed method is termed as DM based SBL (DM-SBL). Numerical simulations demonstrate that DM-SBL significantly outperforms conventional approaches that deal with the AWGN scenario, particularly under low signal-to-interference ratio (SIR) conditions. Beyond channel estimation, DM-SBL also shows promise for addressing other linear inverse problems involving structured interference.

Based on neural network and adaptive subspace approximation method, we propose a new machine learning method for solving partial differential equations. The neural network is adopted to build the basis of the finite dimensional subspace. Then the discrete solution is obtained by using the subspace approximation. Especially, based on the subspace approximation, a posteriori error estimator can be derivated by the hypercircle technique. This a posteriori error estimator can act as the loss function for adaptively refining the parameters of neural network., Comment: 34 pages, 14 figures

We study an $O(N)$ invariant surface defect in the Wilson-Fisher conformal field theory (CFT) in $d=4-\epsilon$ dimensions. This defect is defined by mass deformation on a two-dimensional surface that generates localized disorder and is conjectured to factorize into a pair of ordinary boundary conditions in $d=3$. We determine defect CFT data associated with the lightest $O(N)$ singlet and vector operators up to the third order in the $\epsilon$-expansion, find agreements with results from numerical methods and provide support for the factorization proposal in $d=3$. Along the way, we observe surprising non-renormalization properties for surface anomalous dimensions and operator-product-expansion coefficients in the $\epsilon$-expansion. We also analyze the full conformal anomalies for the surface defect., Comment: 48 pages, 9 figures

The aim of this paper is to introduce a FieldTNN-based machine learning method for solving the Maxwell eigenvalue problem in both 2D and 3D domains, including both tensor and non-tensor computational regions. First, we extend the existing TNN-based approach to address the Maxwell eigenvalue problem, a fundamental challenge in electromagnetic field theory. Second, we tackle non-tensor computational domains, which represents a novel and significant contribution of this work. Third, we incorporate the divergence-free condition into the optimization process, allowing for the automatic filtering of spurious eigenpairs. Numerical examples are presented to demonstrate the efficiency and accuracy of our algorithm, underscoring its potential for broader applications in computational electromagnetics.

Text logo design heavily relies on the creativity and expertise of professional designers, in which arranging element layouts is one of the most important procedures. However, few attention has been paid to this specific task which needs to take precise textural details and user constraints into consideration, but only on the broader tasks such as document/poster layout generation. In this paper, we propose a VLM-based framework that generates content-aware text logo layouts by integrating multi-modal inputs with user constraints, supporting a more flexible and stable layout design in real-world applications. We introduce two model techniques to reduce the computation for processing multiple glyph images simultaneously, while does not face performance degradation. To support instruction-tuning of out model, we construct two extensive text logo datasets, which are 5x more larger than the existing public dataset. Except for the geometric annotations (e.g. text masks and character recognition), we also compliment with comprehensive layout descriptions in natural language format, for more effective training to have reasoning ability when dealing with complex layouts and custom user constraints. Experimental studies demonstrate the effectiveness of our proposed model and datasets, when comparing with previous methods in various benchmarks to evaluate geometric aesthetics and human preferences. The code and datasets will be publicly available.

Crystalline topological insulators and superconductors have been a prominent topic in the field of condensed matter physics. These systems obey certain crystalline (spatial) symmetries that depend on the geometry of the lattice. The presence of spatial symmetries can lead to shift in the classification of ten fold Altland Zirnbauer class, given rise to new symmetry protected topological phases. If the constraint of Hermiticity is broken, the classification expand into 38 fold. In this paper, following procedures in Hermitian systems, we classify all possible types of order-two spatial symmetries for point gapped non Hermitian systems within 16 out of 38 non Hermitian topological classes. These 16 classes are denoted by AZ and AZ^$\dag$ classes. We show that, similar to the Hermitian case, spatial symmetries will also lead to a shift in the classification of AZ and AZ^$\dag$ classes. There also exist novel symmetry protected topological phases exclusive to point gapped non Hermitian Hamiltonians. Toy models are also given based on our classifications.

Monocular 3D object detection has attracted great attention due to simplicity and low cost. Existing methods typically follow conventional 2D detection paradigms, first locating object centers and then predicting 3D attributes via neighboring features. However, these methods predominantly rely on progressive cross-scale feature aggregation and focus solely on local information, which may result in a lack of global awareness and the omission of small-scale objects. In addition, due to large variation in object scales across different scenes and depths, inaccurate receptive fields often lead to background noise and degraded feature representation. To address these issues, we introduces MonoASRH, a novel monocular 3D detection framework composed of Efficient Hybrid Feature Aggregation Module (EH-FAM) and Adaptive Scale-Aware 3D Regression Head (ASRH). Specifically, EH-FAM employs multi-head attention with a global receptive field to extract semantic features for small-scale objects and leverages lightweight convolutional modules to efficiently aggregate visual features across different scales. The ASRH encodes 2D bounding box dimensions and then fuses scale features with the semantic features aggregated by EH-FAM through a scale-semantic feature fusion module. The scale-semantic feature fusion module guides ASRH in learning dynamic receptive field offsets, incorporating scale priors into 3D position prediction for better scale-awareness. Extensive experiments on the KITTI and Waymo datasets demonstrate that MonoASRH achieves state-of-the-art performance.

Jogs, atomic-scale steps on dislocations, play an important role in crystal plasticity, yet they are often ignored in discrete dislocation dynamics (DDD) simulations due to their small sizes. While jogs on screw dislocations are known to move non-conservatively (i.e. climb) accompanied by vacancy emission, jogs on edge dislocations are commonly expected to move conservatively (i.e. glide) with the dislocation under ambient conditions. Here we report unexpected findings from molecular dynamics simulations of an edge dislocation containing a pair of unit jogs in face-centered cubic nickel at 300K. While the jogs glide conservatively with the edge dislocation at low stresses, we observe that one of the jogs climbs and emits vacancies intermittently at higher stresses. This observation is unexpected at such a low temperature, as climb is typically associated with temperatures closer to the creep temperature (roughly half of the melting temperature). Our results highlight the significance of the complex interplay between point defects (i.e., vacancies) and dislocations in room-temperature plasticity, suggesting that these interactions may be more significant than previously thought.

Perspective projection has been extensively utilized in monocular 3D object detection methods. It introduces geometric priors from 2D bounding boxes and 3D object dimensions to reduce the uncertainty of depth estimation. However, due to depth errors originating from the object's visual surface, the height of the bounding box often fails to represent the actual projected central height, which undermines the effectiveness of geometric depth. Direct prediction for the projected height unavoidably results in a loss of 2D priors, while multi-depth prediction with complex branches does not fully leverage geometric depth. This paper presents a Transformer-based monocular 3D object detection method called MonoDGP, which adopts perspective-invariant geometry errors to modify the projection formula. We also try to systematically discuss and explain the mechanisms and efficacy behind geometry errors, which serve as a simple but effective alternative to multi-depth prediction. Additionally, MonoDGP decouples the depth-guided decoder and constructs a 2D decoder only dependent on visual features, providing 2D priors and initializing object queries without the disturbance of 3D detection. To further optimize and fine-tune input tokens of the transformer decoder, we also introduce a Region Segment Head (RSH) that generates enhanced features and segment embeddings. Our monocular method demonstrates state-of-the-art performance on the KITTI benchmark without extra data. Code is available at https://github.com/PuFanqi23/MonoDGP.

Despite significant advancements in natural language generation, controlling language models to produce texts with desired attributes remains a formidable challenge. In this work, we introduce RSA-Control, a training-free controllable text generation framework grounded in pragmatics. RSA-Control directs the generation process by recursively reasoning between imaginary speakers and listeners, enhancing the likelihood that target attributes are correctly interpreted by listeners amidst distractors. Additionally, we introduce a self-adjustable rationality parameter, which allows for automatic adjustment of control strength based on context. Our experiments, conducted with two task types and two types of language models, demonstrate that RSA-Control achieves strong attribute control while maintaining language fluency and content consistency. Our code is available at https://github.com/Ewanwong/RSA-Control., Comment: Accepted to EMNLP 2024 (main conference)

Traditional clustering algorithms often focus on the most fine-grained information and achieve clustering by calculating the distance between each pair of data points or implementing other calculations based on points. This way is not inconsistent with the cognitive mechanism of "global precedence" in human brain, resulting in those methods' bad performance in efficiency, generalization ability and robustness. To address this problem, we propose a new clustering algorithm called granular-ball clustering (GBCT) via granular-ball computing. Firstly, GBCT generates a smaller number of granular-balls to represent the original data, and forms clusters according to the relationship between granular-balls, instead of the traditional point relationship. At the same time, its coarse-grained characteristics are not susceptible to noise, and the algorithm is efficient and robust; besides, as granular-balls can fit various complex data, GBCT performs much better in non-spherical data sets than other traditional clustering methods. The completely new coarse granularity representation method of GBCT and cluster formation mode can also used to improve other traditional methods.

Humans effortlessly integrate common-sense knowledge with sensory input from vision and touch to understand their surroundings. Emulating this capability, we introduce FusionSense, a novel 3D reconstruction framework that enables robots to fuse priors from foundation models with highly sparse observations from vision and tactile sensors. FusionSense addresses three key challenges: (i) How can robots efficiently acquire robust global shape information about the surrounding scene and objects? (ii) How can robots strategically select touch points on the object using geometric and common-sense priors? (iii) How can partial observations such as tactile signals improve the overall representation of the object? Our framework employs 3D Gaussian Splatting as a core representation and incorporates a hierarchical optimization strategy involving global structure construction, object visual hull pruning and local geometric constraints. This advancement results in fast and robust perception in environments with traditionally challenging objects that are transparent, reflective, or dark, enabling more downstream manipulation or navigation tasks. Experiments on real-world data suggest that our framework outperforms previously state-of-the-art sparse-view methods. All code and data are open-sourced on the project website.

Dark-field X-ray microscopy (DFXM) is a novel X-ray imaging technique developed at synchrotrons to image along the diffracted beam with a real space resolution of ~100 nm and reciprocal space resolution of $10^{-4}$. Recent implementations of DFXM at X-ray free electron lasers (XFELs) have demonstrated DFXM's ability to visualize the real-time evolution of coherent GHz phonons produced by ultrafast laser excitation of metal transducers. Combining this with DFXM's ability to visualize strain fields due to dislocations makes it possible to study the interaction of GHz coherent phonons with the strain fields of dislocations, along with studying the damping of coherent phonons due to interactions with thermal phonons. For quantitative analysis of phonon-dislocation interactions and phonon damping, a formalism is required to relate phonon dynamics to the strains measured by DFXM. In this work, we use kinematic diffraction theory to simulate DFXM images of the specific coherent phonons in diamond that are generated by the ultrafast laser excitation of a metal transducer. We extend this formalism to also describe imaging of incoherent phonons of sufficiently high frequency, which are relevant for thermal transport, offering future opportunities for DFXM to image signals produced by thermal diffuse scattering. For both coherent and incoherent phonons, we discuss the optimal sampling of real space, reciprocal space and time, and the opportunities offered by the advances in DFXM optics.

Radio map construction based on extensive measurements is accurate but expensive and time-consuming, while environment-aware radio map estimation reduces the costs at the expense of low accuracy. Considering accuracy and costs, a first-predict-then-correct (FPTC) method is proposed by leveraging generative adversarial networks (GANs). A primary radio map is first predicted by a radio map prediction GAN (RMP-GAN) taking environmental information as input. Then, the prediction result is corrected by a radio map correction GAN (RMC-GAN) with sparse measurements as guidelines. Specifically, the self-attention mechanism and residual-connection blocks are introduced to RMP-GAN and RMC-GAN to improve the accuracy, respectively. Experimental results validate that the proposed FPTC-GANs method achieves the best radio map construction performance, compared with the state-of-the-art methods.

Wireless networks supporting artificial intelligence have gained significant attention, with Over-the-Air Federated Learning emerging as a key application due to its unique transmission and distributed computing characteristics. This paper derives error bounds for Over-the-Air Federated Learning in a Cell-free MIMO system and formulates an optimization problem to minimize optimality gap via joint optimization of power control and beamforming. We introduce the MOP-LOFPC algorithm, which employs Lyapunov optimization to decouple long-term constraints across rounds while requiring only causal channel state information. Experimental results demonstrate that MOP-LOFPC achieves a better and more flexible trade-off between the model's training loss and adherence to long-term power constraints compared to existing baselines.

Non-intrusive Appliance Load Monitoring (NALM) aims to recognize individual appliance usage from the main meter without indoor sensors. However, existing systems struggle to balance dataset construction efficiency and event/state recognition accuracy, especially for low-power appliance recognition. This paper introduces Hawk, an efficient and accurate NALM system that operates in two stages: dataset construction and event recognition. In the data construction stage, we efficiently collect a balanced and diverse dataset, HawkDATA, based on balanced Gray code and enable automatic data annotations via a sampling synchronization strategy called shared perceptible time. During the event recognition stage, our algorithm integrates steady-state differential pre-processing and voting-based post-processing for accurate event recognition from the aggregate current. Experimental results show that HawkDATA takes only 1/71.5 of the collection time to collect 6.34x more appliance state combinations than the baseline. In HawkDATA and a widely used dataset, Hawk achieves an average F1 score of 93.94% for state recognition and 97.07% for event recognition, which is a 47. 98% and 11. 57% increase over SOTA algorithms. Furthermore, selected appliance subsets and the model trained from HawkDATA are deployed in two real-world scenarios with many unknown background appliances. The average F1 scores of event recognition are 96.02% and 94.76%. Hawk's source code and HawkDATA are accessible at https://github.com/WZiJ/SenSys24-Hawk., Comment: Accepted to the 22nd ACM Conference on Embedded Networked Sensor Systems (SenSys 2024)

Reconstructing urban street scenes is crucial due to its vital role in applications such as autonomous driving and urban planning. These scenes are characterized by long and narrow camera trajectories, occlusion, complex object relationships, and data sparsity across multiple scales. Despite recent advancements, existing surface reconstruction methods, which are primarily designed for object-centric scenarios, struggle to adapt effectively to the unique characteristics of street scenes. To address this challenge, we introduce StreetSurfGS, the first method to employ Gaussian Splatting specifically tailored for scalable urban street scene surface reconstruction. StreetSurfGS utilizes a planar-based octree representation and segmented training to reduce memory costs, accommodate unique camera characteristics, and ensure scalability. Additionally, to mitigate depth inaccuracies caused by object overlap, we propose a guided smoothing strategy within regularization to eliminate inaccurate boundary points and outliers. Furthermore, to address sparse views and multi-scale challenges, we use a dual-step matching strategy that leverages adjacent and long-term information. Extensive experiments validate the efficacy of StreetSurfGS in both novel view synthesis and surface reconstruction.

Given an input painting, we reconstruct a time-lapse video of how it may have been painted. We formulate this as an autoregressive image generation problem, in which an initially blank "canvas" is iteratively updated. The model learns from real artists by training on many painting videos. Our approach incorporates text and region understanding to define a set of painting "instructions" and updates the canvas with a novel diffusion-based renderer. The method extrapolates beyond the limited, acrylic style paintings on which it has been trained, showing plausible results for a wide range of artistic styles and genres., Comment: Project Page: https://inversepainting.github.io

In this paper, we introduce an adapted one-dimensional (1D) reduced model aimed at analyzing blood flow within stenosed arteries. Differing from the prevailing 1D model \cite{Formaggia2003, Sherwin2003_2, Sherwin2003, Quarteroni2004, 10.1007/978-3-642-56288-4_10}, our approach incorporates the variable radius of the blood vessel. Our methodology begins with the non-dimensionalization of the Navier-Stokes equations for axially symmetric flow in cylindrical coordinates and then derives the extended 1D reduced model, by making additional adjustments to accommodate the effects of variable radii of the vessel along the longitudinal direction. Additionally, we propose a method to extract radial velocity information from the 1D results during post-processing, enabling the generation of two-dimensional (2D) velocity data. We validate our model by conducting numerical simulations of blood flow through stenotic arteries with varying severities, ranging from 23% to 50%. The results were compared to those from the established 1D model and a full three-dimensional (3D) simulation, highlighting the potential and importance of this model for arteries with variable radius. All the code used to generate the results presented in the paper is available at https://github.com/qcutexu/Extended-1D-AQ-system.git.

We report the fabrication and characteristics of GaAs/Si p+/n+ heterojunction tunnel diodes. These diodes were fabricated via grafting the freestanding single-crystalline p-type degenerately doped GaAs (4E19 cm-3) nanomembrane (NM) onto single-crystalline n-type Si (5E19 cm-3) substrate. At the heterointerface, an amorphous ultrathin oxygen-enriched layer (UOL) was intentionally engineered through chemical oxidation and atomic layer deposition (ALD). Scanning transmission electron microscopy (STEM) confirmed the formation of the UOL and the single crystallinity of the grafted junction. The resulting tunnel diodes consistently exhibited negative differential resistance (NDR) behavior at room temperature, with a high maximum peak-to-valley current ratio (PVCR) of 36.38, valley voltages ranging from 1.3 to 1.8 V, and a peak tunneling current density of 0.95 kA/cm2. This study not only highlights the critical roles of the UOL as both an interface improvement layer and a quantum tunneling medium, but also establishes "semiconductor grafting" as an effective and versatile method for high-performance, lattice-mismatched heterojunction devices., Comment: 4 pages, 5 figures

Finger vein recognition technology has become one of the primary solutions for high-security identification systems. However, it still has information leakage problems, which seriously jeopardizes users privacy and anonymity and cause great security risks. In addition, there is no work to consider a fully integrated secure finger vein recognition system. So, different from the previous systems, we integrate preprocessing and template protection into an integrated deep learning model. We propose an end-to-end cancelable finger vein network (CFVNet), which can be used to design an secure finger vein recognition system.It includes a plug-and-play BWR-ROIAlign unit, which consists of three sub-modules: Localization, Compression and Transformation. The localization module achieves automated localization of stable and unique finger vein ROI. The compression module losslessly removes spatial and channel redundancies. The transformation module uses the proposed BWR method to introduce unlinkability, irreversibility and revocability to the system. BWR-ROIAlign can directly plug into the model to introduce the above features for DCNN-based finger vein recognition systems. We perform extensive experiments on four public datasets to study the performance and cancelable biometric attributes of the CFVNet-based recognition system. The average accuracy, EERs and Dsys on the four datasets are 99.82%, 0.01% and 0.025, respectively, and achieves competitive performance compared with the state-of-the-arts.

We investigate the fluid-poroelastic structure interaction problem in a moving domain, governed by Navier-Stokes-Biot (NSBiot) system. First, we propose a fully parallelizable, loosely coupled scheme to solve the coupled system. At each time step, the solution from the previous time step is used to approximate the coupling conditions at the interface, allowing the original coupled problem to be fully decoupled into seperate fluid and structure subproblems, which are solved in parallel. Since our approach utilizes a loosely coupled scheme, no sub-iterations are required at each time step. Next, we conduct the energy estimates of this splitting method for the linearized problem (Stokes-Biot system), which demonstrates that the scheme is unconditionally stable without any restriction of the time step size from the physical parameters. Furthermore, we illustrate the first-order accuracy in time through two benchmark problems. Finally, to demonstrate that the proposed method maintains its excellent stability properties also for the nonlinear NSBiot system, we present numerical results for both $2D$ and $3D$ NSBiot problems related to real-world physical applications.

The growing demand for AI training data has transformed data annotation into a global industry, but traditional approaches relying on human annotators are often time-consuming, labor-intensive, and prone to inconsistent quality. We propose the Model-in-the-Loop (MILO) framework, which integrates AI/ML models into the annotation process. Our research introduces a collaborative paradigm that leverages the strengths of both professional human annotators and large language models (LLMs). By employing LLMs as pre-annotation and real-time assistants, and judges on annotator responses, MILO enables effective interaction patterns between human annotators and LLMs. Three empirical studies on multimodal data annotation demonstrate MILO's efficacy in reducing handling time, improving data quality, and enhancing annotator experiences. We also introduce quality rubrics for flexible evaluation and fine-grained feedback on open-ended annotations. The MILO framework has implications for accelerating AI/ML development, reducing reliance on human annotation alone, and promoting better alignment between human and machine values.

Machine-learned force fields (MLFFs) promise to offer a computationally efficient alternative to ab initio simulations for complex molecular systems. However, ensuring their generalizability beyond training data is crucial for their wide application in studying solid materials. This work investigates the ability of a graph neural network (GNN)-based MLFF, trained on Lennard-Jones Argon, to describe solid-state phenomena not explicitly included during training. We assess the MLFF's performance in predicting phonon density of states (PDOS) for a perfect face-centered cubic (FCC) crystal structure at both zero and finite temperatures. Additionally, we evaluate vacancy migration rates and energy barriers in an imperfect crystal using direct molecular dynamics (MD) simulations and the string method. Notably, vacancy configurations were absent from the training data. Our results demonstrate the MLFF's capability to capture essential solid-state properties with good agreement to reference data, even for unseen configurations. We further discuss data engineering strategies to enhance the generalizability of MLFFs. The proposed set of benchmark tests and workflow for evaluating MLFF performance in describing perfect and imperfect crystals pave the way for reliable application of MLFFs in studying complex solid-state materials., Comment: 17 pages, 7 figures

3D Gaussian Splatting (3DGS) has shown promising performance in novel view synthesis. Previous methods adapt it to obtaining surfaces of either individual 3D objects or within limited scenes. In this paper, we make the first attempt to tackle the challenging task of large-scale scene surface reconstruction. This task is particularly difficult due to the high GPU memory consumption, different levels of details for geometric representation, and noticeable inconsistencies in appearance. To this end, we propose GigaGS, the first work for high-quality surface reconstruction for large-scale scenes using 3DGS. GigaGS first applies a partitioning strategy based on the mutual visibility of spatial regions, which effectively grouping cameras for parallel processing. To enhance the quality of the surface, we also propose novel multi-view photometric and geometric consistency constraints based on Level-of-Detail representation. In doing so, our method can reconstruct detailed surface structures. Comprehensive experiments are conducted on various datasets. The consistent improvement demonstrates the superiority of GigaGS.

In their comment (1), Hawreliak et al. claims that our observation of stacking fault formation and transonic dislocation propagation in diamond (2) is not valid as they interpret the observed features as cracks. In this response letter, we describe our rationale for interpreting the observed features as stacking faults. We also address other points raised in their comments, including the clarifications of how the results of Makarov et al. (3) are not in conflict with our study., Comment: Comment on arXiv:2401.04213. 10 pages, 2 figures

Fine-grained simulation of floor construction processes is essential for supporting lean management and the integration of information technology. However, existing research does not adequately address the on-site decision-making of constructors in selecting tasks and determining their sequence within the entire construction process. Moreover, decision-making frameworks from computer science and robotics are not directly applicable to construction scenarios. To facilitate intelligent simulation in construction, this study introduces the Construction Markov Decision Process (CMDP). The primary contribution of this CMDP framework lies in its construction knowledge in decision, observation modifications and policy design, enabling agents to perceive the construction state and follow policy guidance to evaluate and reach various range of targets for optimizing the planning of construction activities. The CMDP is developed on the Unity platform, utilizing a two-stage training approach with the multi-agent proximal policy optimization algorithm. A case study demonstrates the effectiveness of this framework: the low-level policy successfully simulates the construction process in continuous space, facilitating policy testing and training focused on reducing conflicts and blockages among crews; and the high-level policy improving the spatio-temporal planning of construction activities, generating construction patterns in distinct phases, leading to the discovery of new construction insights.

Dark-field X-ray Microscopy (DFXM) is a novel diffraction-based imaging technique that non-destructively maps the local deformation from crystalline defects in bulk materials. While studies have demonstrated that DFXM can spatially map 3D defect geometries, it is still challenging to interpret DFXM images of the high dislocation density systems relevant to macroscopic crystal plasticity. This work develops a scalable forward model to calculate virtual DFXM images for complex discrete dislocation (DD) structures obtained from atomistic simulations. Our new DD-DFXM model integrates a non-singular formulation for calculating the local strain from the DD structures and an efficient geometrical optics algorithm for computing the DFXM image from the strain. We apply the model to complex DD structures obtained from a large-scale molecular dynamics (MD) simulation of compressive loading on a single-crystal silicon. Simulated DFXM images exhibit prominent feature contrast for dislocations between the multiple slip systems, demonstrating the DFXM's potential to resolve features from dislocation multiplication. The integrated DD-DFXM model provides a toolbox for DFXM experimental design and image interpretation in the context of bulk crystal plasticity for the breadth of measurements across shock plasticity and the broader materials science community.

Internet Service Providers (ISPs) bear the brunt of being the first port of call for poor video streaming experience. ISPs can benefit from knowing the user's device type (e.g., Android, iOS) and software agent (e.g., native app, Chrome) to troubleshoot platform-specific issues, plan capacity and create custom bundles. Unfortunately, encryption and NAT have limited ISPs' visibility into user platforms across video streaming providers. We develop a methodology to identify user platforms for video streams from four popular providers, namely YouTube, Netflix, Disney, and Amazon, by analyzing network traffic in real-time. First, we study the anatomy of the connection establishment process to show how TCP/QUIC and TLS handshakes vary across user platforms. We then develop a classification pipeline that uses 62 attributes extracted from the handshake messages to determine the user device and software agent of video flows with over 96% accuracy. Our method is evaluated and deployed in a large campus network (mimicking a residential broadband network) serving users including dormitory residents. Analysis of 100+ million video streams over a four-month period reveals insights into the mix of user platforms across the video providers, variations in bandwidth consumption across operating systems and browsers, and differences in peak hours of usage., Comment: This paper is accepted at ACM Internet Measurement Conference (IMC) 2024. Proc. ACM IMC, Madrid, Spain, Nov 2024

Neural implicit reconstruction via volume rendering has demonstrated its effectiveness in recovering dense 3D surfaces. However, it is non-trivial to simultaneously recover meticulous geometry and preserve smoothness across regions with differing characteristics. To address this issue, previous methods typically employ geometric priors, which are often constrained by the performance of the prior models. In this paper, we propose ND-SDF, which learns a Normal Deflection field to represent the angular deviation between the scene normal and the prior normal. Unlike previous methods that uniformly apply geometric priors on all samples, introducing significant bias in accuracy, our proposed normal deflection field dynamically learns and adapts the utilization of samples based on their specific characteristics, thereby improving both the accuracy and effectiveness of the model. Our method not only obtains smooth weakly textured regions such as walls and floors but also preserves the geometric details of complex structures. In addition, we introduce a novel ray sampling strategy based on the deflection angle to facilitate the unbiased rendering process, which significantly improves the quality and accuracy of intricate surfaces, especially on thin structures. Consistent improvements on various challenging datasets demonstrate the superiority of our method.

Graph neural networks (GNNs) have achieved impressive performance in graph domain adaptation. However, extensive source graphs could be unavailable in real-world scenarios due to privacy and storage concerns. To this end, we investigate an underexplored yet practical problem of source-free graph domain adaptation, which transfers knowledge from source models instead of source graphs to a target domain. To solve this problem, we introduce a novel GNN-based approach called Rank and Align (RNA), which ranks graph similarities with spectral seriation for robust semantics learning, and aligns inharmonic graphs with harmonic graphs which close to the source domain for subgraph extraction. In particular, to overcome label scarcity, we employ the spectral seriation algorithm to infer the robust pairwise rankings, which can guide semantic learning using a similarity learning objective. To depict distribution shifts, we utilize spectral clustering and the silhouette coefficient to detect harmonic graphs, which the source model can easily classify. To reduce potential domain discrepancy, we extract domain-invariant subgraphs from inharmonic graphs by an adversarial edge sampling process, which guides the invariant learning of GNNs. Extensive experiments on several benchmark datasets demonstrate the effectiveness of our proposed RNA., Comment: Published in IJCAI2024

Signed Distance Function (SDF)-based volume rendering has demonstrated significant capabilities in surface reconstruction. Although promising, SDF-based methods often fail to capture detailed geometric structures, resulting in visible defects. By comparing SDF-based volume rendering to density-based volume rendering, we identify two main factors within the SDF-based approach that degrade surface quality: SDF-to-density representation and geometric regularization. These factors introduce challenges that hinder the optimization of the SDF field. To address these issues, we introduce NeuRodin, a novel two-stage neural surface reconstruction framework that not only achieves high-fidelity surface reconstruction but also retains the flexible optimization characteristics of density-based methods. NeuRodin incorporates innovative strategies that facilitate transformation of arbitrary topologies and reduce artifacts associated with density bias. Extensive evaluations on the Tanks and Temples and ScanNet++ datasets demonstrate the superiority of NeuRodin, showing strong reconstruction capabilities for both indoor and outdoor environments using solely posed RGB captures. Project website: https://open3dvlab.github.io/NeuRodin/

Nanoparticle surface structural dynamics is believed to play a significant role in regulating functionalities such as diffusion, reactivity, and catalysis but the atomic-level processes are not well understood. Atomic resolution characterization of nanoparticle surface dynamics is challenging since it requires both high spatial and temporal resolution. Though ultrafast transmission electron microscopy (TEM) can achieve picosecond temporal resolution, it is limited to nanometer spatial resolution. On the other hand, with the high readout rate of new electron detectors, conventional TEM has the potential to visualize atomic structure with millisecond time resolutions. However, the need to limit electron dose rates to reduce beam damage yields millisecond images that are dominated by noise, obscuring structural details. Here we show that a newly developed unsupervised denoising framework based on artificial intelligence enables observations of metal nanoparticle surfaces with time resolutions down to 10 ms at moderate electron dose. On this timescale, we find that many nanoparticle surfaces continuously transition between ordered and disordered configurations. The associated stress fields can penetrate below the surface leading to defect formation and destabilization making the entire nanoparticle fluxional. Combining this unsupervised denoiser with electron microscopy greatly improves spatio-temporal characterization capabilities, opening a new window for future exploration of atomic-level structural dynamics in materials.

Most multi-agent reinforcement learning approaches adopt two types of policy optimization methods that either update policy simultaneously or sequentially. Simultaneously updating policies of all agents introduces non-stationarity problem. Although sequentially updating policies agent-by-agent in an appropriate order improves policy performance, it is prone to low efficiency due to sequential execution, resulting in longer model training and execution time. Intuitively, partitioning policies of all agents according to their interdependence and updating joint policy batch-by-batch can effectively balance performance and efficiency. However, how to determine the optimal batch partition of policies and batch updating order are challenging problems. Firstly, a sequential batched policy updating scheme, B2MAPO (Batch by Batch Multi-Agent Policy Optimization), is proposed with a theoretical guarantee of the monotonic incrementally tightened bound. Secondly, a universal modulized plug-and-play B2MAPO hierarchical framework, which satisfies CTDE principle, is designed to conveniently integrate any MARL models to fully exploit and merge their merits, including policy optimality and inference efficiency. Next, a DAG-based B2MAPO algorithm is devised, which is a carefully designed implementation of B2MAPO framework. Comprehensive experimental results conducted on StarCraftII Multi-agent Challenge and Google Football Research demonstrate the performance of DAG-based B2MAPO algorithm outperforms baseline methods. Meanwhile, compared with A2PO, our algorithm reduces the model training and execution time by 60.4% and 78.7%, respectively.

Graph classification is a critical task in numerous multimedia applications, where graphs are employed to represent diverse types of multimedia data, including images, videos, and social networks. Nevertheless, in real-world scenarios, labeled graph data can be limited or scarce. To address this issue, we focus on the problem of semi-supervised graph classification, which involves both supervised and unsupervised models learning from labeled and unlabeled data. In contrast to recent approaches that transfer the entire knowledge from the unsupervised model to the supervised one, we argue that an effective transfer should only retain the relevant semantics that align well with the supervised task. In this paper, we propose a novel framework named DisenSemi, which learns disentangled representation for semi-supervised graph classification. Specifically, a disentangled graph encoder is proposed to generate factor-wise graph representations for both supervised and unsupervised models. Then we train two models via supervised objective and mutual information (MI)-based constraints respectively. To ensure the meaningful transfer of knowledge from the unsupervised encoder to the supervised one, we further define an MI-based disentangled consistency regularization between two models and identify the corresponding rationale that aligns well with the current graph classification task. Experimental results on a range of publicly accessible datasets reveal the effectiveness of our DisenSemi., Comment: Accepted by IEEE Transactions on Neural Networks and Learning Systems (TNNLS 2024)

Electron-impact ionization cross sections of atoms and molecules are essential for plasma modelling. However, experimentally determining the absolute cross sections is not easy, and ab initio calculations become computationally prohibitive as molecular complexity increases. Existing AI-based prediction methods suffer from limited data availability and poor generalization. To address these issues, we propose DeepCSNet, a deep learning approach designed to predict electron-impact ionization cross sections using limited training data. We present two configurations of DeepCSNet: one tailored for specific molecules and another for various molecules. Both configurations can typically achieve a relative L2 error less than 5%. The present numerical results, focusing on electron-impact doubly differential ionization cross sections, demonstrate DeepCSNet's generalization ability, predicting cross sections across a wide range of energies and incident angles. Additionally, DeepCSNet shows promising results in predicting cross sections for molecules not included in the training set, even large molecules with more than 10 constituent atoms, highlighting its potential for practical applications., Comment: 18 pages, 16 figures, 2 tables

Capacitive Micromachined Ultrasonic Transducer (CMUT) has a wild range of applications in medical detecting and imaging fields. However, operating under self-generating-self-receiving (SGSR) method usually results in poor sensitivity. But the sensitivity cannot be improved simply by increasing the resonant frequency since the frequency of a specific kind of CMUT is designed for specific usage. In this paper, based on one specific type of CMUT, mechanical model is built and simulation analysis is demonstrated. A brand-new method one-generating-multiple-receiving (OGMR) is introduced and a special circuit model has been designed to improve the signal to thermal noise ratio. By increasing the number of receiving capacitors from 1 to 8, we increased the signal-noise ratio to 2.83 times., Comment: 6 pages, 18 figures

Task-specific functional MRI (fMRI) images provide excellent modalities for studying the neuronal basis of cognitive processes. We use fMRI data to formulate and solve the problem of deconvolving task-specific aggregate neuronal networks into a set of basic building blocks called canonical networks, to use these networks for functional characterization, and to characterize the physiological basis of these responses by mapping them to regions of the brain. Our results show excellent task-specificity of canonical networks, i.e., the expression of a small number of canonical networks can be used to accurately predict tasks; generalizability across cohorts, i.e., canonical networks are conserved across diverse populations, studies, and acquisition protocols; and that canonical networks have strong anatomical and physiological basis. From a methods perspective, the problem of identifying these canonical networks poses challenges rooted in the high dimensionality, small sample size, acquisition variability, and noise. Our deconvolution technique is based on non-negative matrix factorization (NMF) that identifies canonical networks as factors of a suitably constructed matrix. We demonstrate that our method scales to large datasets, yields stable and accurate factors, and is robust to noise., Comment: 9 pages, 5 figures

Aligning large language models (LLMs) with human preferences is critical for their deployment. Recently, decoding-time alignment has emerged as an effective plug-and-play technique that requires no fine-tuning of model parameters. However, generating text that achieves both high reward and high likelihood remains a significant challenge. Existing methods often fail to generate high-reward text or incur substantial computational costs. In this paper, we propose Cascade Reward Sampling (CARDS) to address both issues, guaranteeing the generation of high-reward and high-likelihood text with significantly low costs. Based on our analysis of reward models (RMs) on incomplete text and our observation that high-reward prefixes induce high-reward complete text, we use rejection sampling to iteratively generate small semantic segments to form such prefixes. The segment length is dynamically determined by the predictive uncertainty of LLMs. This strategy guarantees desirable prefixes for subsequent generations and significantly reduces wasteful token re-generations and the number of reward model scoring. Our experiments demonstrate substantial gains in both generation efficiency and alignment ratings compared to the baselines, achieving five times faster text generation and 99\% win-ties in GPT-4/Claude-3 helpfulness evaluation.

Large language models (LLMs) have revolutionized lots of fields of research. Although it is well-known that fine-tuning is essential for enhancing the capabilities of LLMs, existing research suggests that there is potential redundancy in the fine-tuning process and therefore proposes to update only a subset of parameters. However, these methods fail to leverage the task-specific information to identify important parameters during training. Based on the insight that gradients inherently contain information on task-specific data, we propose Gradient-Mask Tuning (GMT), a method that selectively updates parameters during training based on their gradient information. Specifically, we compute the absolute values of the gradients and apply masking to those with relatively smaller magnitudes. Our empirical results across various tasks demonstrate that GMT not only outperforms traditional fine-tuning methods but also elevates the upper limits of LLM performance. Further analysis indicates that GMT exhibits insensitivity to mask ratio and possesses computational efficiency comparable to vanilla SFT.

This paper introduces a tensor neural network (TNN) to address nonparametric regression problems, leveraging its distinct sub-network structure to effectively facilitate variable separation and enhance the approximation of complex, high-dimensional functions. The TNN demonstrates superior performance compared to conventional Feed-Forward Networks (FFN) and Radial Basis Function Networks (RBN) in terms of both approximation accuracy and generalization capacity, even with a comparable number of parameters. A significant innovation in our approach is the integration of statistical regression and numerical integration within the TNN framework. This allows for efficient computation of high-dimensional integrals associated with the regression function and provides detailed insights into the underlying data structure. Furthermore, we employ gradient and Laplacian analysis on the regression outputs to identify key dimensions influencing the predictions, thereby guiding the design of subsequent experiments. These advancements make TNN a powerful tool for applications requiring precise high-dimensional data analysis and predictive modeling.

Despite the superior capabilities of Multimodal Large Language Models (MLLMs) across diverse tasks, they still face significant trustworthiness challenges. Yet, current literature on the assessment of trustworthy MLLMs remains limited, lacking a holistic evaluation to offer thorough insights into future improvements. In this work, we establish MultiTrust, the first comprehensive and unified benchmark on the trustworthiness of MLLMs across five primary aspects: truthfulness, safety, robustness, fairness, and privacy. Our benchmark employs a rigorous evaluation strategy that addresses both multimodal risks and cross-modal impacts, encompassing 32 diverse tasks with self-curated datasets. Extensive experiments with 21 modern MLLMs reveal some previously unexplored trustworthiness issues and risks, highlighting the complexities introduced by the multimodality and underscoring the necessity for advanced methodologies to enhance their reliability. For instance, typical proprietary models still struggle with the perception of visually confusing images and are vulnerable to multimodal jailbreaking and adversarial attacks; MLLMs are more inclined to disclose privacy in text and reveal ideological and cultural biases even when paired with irrelevant images in inference, indicating that the multimodality amplifies the internal risks from base LLMs. Additionally, we release a scalable toolbox for standardized trustworthiness research, aiming to facilitate future advancements in this important field. Code and resources are publicly available at: https://multi-trust.github.io/., Comment: 100 pages, 84 figures, 33 tables