Search

Mobile government (m-government) represents a distinct paradigm shift from electronic government (e-government), offering a new avenue for governments worldwide to deliver services and applications to their customers. The m-government model deviates from e-government in terms of information technology (IT) infrastructure, security, and application management and implementation. Enterprise architecture (EA) has been developed and utilized globally to enhance efficiency and information and communication technology (ICT) utilization in the public sector through e-government. However, the application of EA within the context of m-government, particularly in developing countries, has largely been overlooked by scholars. This study aims to address this gap. This study seeks to develop an EA specifically tailored for m-government in a developmental context. Our contribution to the literature is the illustration of a proposed EA framework for m-government. The practical implementation of this study is to identify critical considerations when designing and adopting m-government to avoid redundant investments during the integration of infrastructure and applications from e-government to m-government.

Transformer-based large-scale pre-trained models achieve great success, and fine-tuning, which tunes a pre-trained model on a task-specific dataset, is the standard practice to utilize these models for downstream tasks. Recent work has developed adapter-tuning, but these approaches either still require a relatively high resource usage. Through our investigation, we show that each adapter in adapter-tuning does not have the same impact on task performance and resource usage. Based on our findings, we propose SAFE, which gradually freezes less-important adapters that do not contribute to adaptation during the early training steps. In our experiments, SAFE reduces memory usage, computation amount, and training time by 42.85\%, 34.59\%, and 11.82\%, respectively, while achieving comparable or better performance compared to the baseline. We also demonstrate that SAFE induces regularization effect, thereby smoothing the loss landscape.

Enhancing light emission from perovskite nanocrystal (NC) films is essential in light-emitting devices, as their conventional stacks often restrict the escape of emitted light. This work addresses this challenge by employing a TiO$_2$ grating to enhance light extraction and shape the emission of CsPbBr$_3$ nanocrystal films. Angle-resolved photoluminescence (PL) demonstrated a tenfold increase in emission intensity by coupling the Bloch resonances of the grating with the spontaneous emission of the perovskite NCs. Fluorescence lifetime imaging microscopy (FLIM) provided micrometer-resolution mapping of both PL intensity and lifetime across a large area, revealing a decrease in PL lifetime from 8.2 ns for NC films on glass to 6.1 ns on the TiO$_2$ grating. Back focal plane (BFP) spectroscopy confirmed how the Bloch resonances transformed the unpolarized, spatially incoherent emission of NCs into polarized and directed light. These findings provide further insights into the interactions between dielectric nanostructures and perovskite NC films, offering possible pathways for designing better performing perovskite optoelectronic devices.

Deep Operator Networks (DeepONets) are among the most prominent frameworks for operator learning, grounded in the universal approximation theorem for operators. However, training DeepONets typically requires significant computational resources. To address this limitation, we propose ELM-DeepONets, an Extreme Learning Machine (ELM) framework for DeepONets that leverages the backpropagation-free nature of ELM. By reformulating DeepONet training as a least-squares problem for newly introduced parameters, the ELM-DeepONet approach significantly reduces training complexity. Validation on benchmark problems, including nonlinear ODEs and PDEs, demonstrates that the proposed method not only achieves superior accuracy but also drastically reduces computational costs. This work offers a scalable and efficient alternative for operator learning in scientific computing.

We present the results from long-term simultaneous monitoring observations of SiO and H2O masers toward the Mira variable star WX Serpentis. This study has been conducted with 21m single-dish radio telescopes of the Korean VLBI Network from 2009 June to 2021 June. Five maser lines were considered: SiO v=1, 2, J=1-0; SiO v=1, J=2-1, 3-2, and H2O 6(1,6)-5(2,3) transitions, with the SiO maser lines distributed near the stellar velocity and the H2O maser exhibiting an asymmetric line profile with five to six peaked components. Intense H2O maser emissions suddenly appeared in 2019 September, indicating flaring. The intensity variations of SiO and H2O masers are strongly correlated with the optical light curve (OLC) of the central star, with individual phase lags; the phase lag of the H2O maser relative to the OLC is larger than that of the SiO masers. The consequent phase difference between the SiO masers and the H2O maser likely indicates that their formation regions and main driving mechanisms are different from each other. The SiO masers in WX Ser exhibit a dominant single-peak velocity distribution, similar to other Mira variable stars. However, the H2O maser displays distinct morphological features, showing a radial acceleration and preferential intensity dominance at blueshifted velocities. This suggests that the H2O maser clouds of WX Ser are moving outward, thereby developing an asymmetric outflow owing to nonuniform material ejection from the stellar atmosphere. The findings confirm that an initial asymmetric outflow structure emerged during the thermally pulsing asymptotic giant branch phase, specifically in the Mira variable star stage., Comment: 24 pages, 10 figures

Automated audio captioning is a task that generates textual descriptions for audio content, and recent studies have explored using visual information to enhance captioning quality. However, current methods often fail to effectively fuse audio and visual data, missing important semantic cues from each modality. To address this, we introduce LAVCap, a large language model (LLM)-based audio-visual captioning framework that effectively integrates visual information with audio to improve audio captioning performance. LAVCap employs an optimal transport-based alignment loss to bridge the modality gap between audio and visual features, enabling more effective semantic extraction. Additionally, we propose an optimal transport attention module that enhances audio-visual fusion using an optimal transport assignment map. Combined with the optimal training strategy, experimental results demonstrate that each component of our framework is effective. LAVCap outperforms existing state-of-the-art methods on the AudioCaps dataset, without relying on large datasets or post-processing. Code is available at https://github.com/NAVER-INTEL-Co-Lab/gaudi-lavcap., Comment: 5 pages, 2 figures; Accepted to ICASSP 2025

We introduce HRMCR (HAE-RAE Multi-Step Commonsense Reasoning), a benchmark designed to evaluate large language models' ability to perform multi-step reasoning in culturally specific contexts, focusing on Korean. The questions are automatically generated via templates and algorithms, requiring LLMs to integrate Korean cultural knowledge into sequential reasoning steps. Consistent with prior observations on emergent abilities, our experiments reveal that models trained on fewer than \(2 \cdot 10^{25}\) training FLOPs struggle to solve any questions, showing near-zero performance. Beyond this threshold, performance improves sharply. State-of-the-art models (e.g., O1) still score under 50\%, underscoring the difficulty of our tasks. Notably, stepwise analysis suggests the observed emergent behavior may stem from compounding errors across multiple steps rather than reflecting a genuinely new capability. We publicly release the benchmark and commit to regularly updating the dataset to prevent contamination., Comment: 11 pages, 7 figures

Urban traffic systems are characterized by dynamic interactions between congestion and free-flow states, influenced by human activity and road topology. This study employs percolation theory to analyze traffic dynamics in Seoul, focusing on the transition point $q_c$ and Fisher exponent $\tau$. The transition point $q_c$ quantifies the robustness of the free-flow clusters, while the exponent $\tau$ captures the spatial fragmentation of the traffic networks. Our analysis reveals temporal variations in these metrics, with lower $q_c$ and lower $\tau$ values during rush hours representing low-dimensional behavior. Weight-weight correlations are found to significantly impact cluster formation, driving the early onset of dominant traffic states. Comparisons with uncorrelated models highlight the role of real-world correlations. This approach provides a comprehensive framework for evaluating traffic resilience and informs strategies to optimize urban transportation systems., Comment: 7 pages, 5 figures

This paper presents the design a Proportional-Integral-Derivative (PID) controller with optimized parameters for a two-degree-of-freedom robotic arm. A genetic algorithm (GA) is proposed to optimize the controller parameters, addressing the challenges in determining PID controller parameters for highly nonlinear systems like robotic arms compared to traditional methods. The GA-optimized PID controller significantly improves control accuracy and performance over traditional control methods. Simulation results demonstrate that the robotic arm system operates with high precision and stability. Additionally, the shortened trajectory tracking response time enhances the feasibility of applying this control algorithm in realworld scenarios. This research not only confirms the suitability of PID-GA for robotic arms and similar systems but also opens new avenues for applying this algorithm to real physical systems.

The profiled vehicle routing problem (PVRP) is a generalization of the heterogeneous capacitated vehicle routing problem (HCVRP) in which the objective is to optimize the routes of vehicles to serve client demands subject to different vehicle profiles, with each having a preference or constraint on a per-client basis. While existing learning methods have shown promise for solving the HCVRP in real-time, no learning method exists to solve the more practical and challenging PVRP. In this paper, we propose a Collaborative Attention Model with Profiles (CAMP), a novel approach that learns efficient solvers for PVRP using multi-agent reinforcement learning. CAMP employs a specialized attention-based encoder architecture to embed profiled client embeddings in parallel for each vehicle profile. We design a communication layer between agents for collaborative decision-making across profiled embeddings at each decoding step and a batched pointer mechanism to attend to the profiled embeddings to evaluate the likelihood of the next actions. We evaluate CAMP on two variants of PVRPs: PVRP with preferences, which explicitly influence the reward function, and PVRP with zone constraints with different numbers of agents and clients, demonstrating that our learned solvers achieve competitive results compared to both classical state-of-the-art neural multi-agent models in terms of solution quality and computational efficiency. We make our code openly available at https://github.com/ai4co/camp., Comment: Accepted at AAMAS 2025

Large language models (LLMs) demonstrate exceptional performance on complex reasoning tasks. However, despite their strong reasoning capabilities in high-resource languages (e.g., English and Chinese), a significant performance gap persists in other languages. To investigate this gap in Korean, we introduce HRM8K, a benchmark comprising 8,011 English-Korean parallel bilingual math problems. Through systematic analysis of model behaviors, we identify a key finding: these performance disparities stem primarily from difficulties in comprehending non-English inputs, rather than limitations in reasoning capabilities. Based on these findings, we propose UST (Understand, Solve, and Translate), a method that strategically uses English as an anchor for reasoning and solution generation. By fine-tuning the model on 130k synthetically generated data points, UST achieves a 10.91% improvement on the HRM8K benchmark and reduces the multilingual performance gap from 11.6% to 0.7%. Additionally, we show that improvements from UST generalize effectively to different Korean domains, demonstrating that capabilities acquired from machine-verifiable content can be generalized to other areas. We publicly release the benchmark, training dataset, and models., Comment: 18 pages, 14 figures, 9 tables

Gait and movement analysis have become a well-established clinical tool for diagnosing health conditions, monitoring disease progression for a wide spectrum of diseases, and to implement and assess treatment, surgery and or rehabilitation interventions. However, quantitative motion assessment remains limited to costly motion capture systems and specialized personnel, restricting its accessibility and broader application. Recent advancements in deep neural networks have enabled quantitative movement analysis using single-camera videos, offering an accessible alternative to conventional motion capture systems. In this paper, we present an efficient approach for clinical gait analysis through a dual-pattern input convolutional Transformer network. The proposed system leverages a dual-input Transformer model to estimate essential gait parameters from single RGB videos captured by a single-view camera. The system demonstrates high accuracy in estimating critical metrics such as the gait deviation index (GDI), knee flexion angle, step length, and walking cadence, validated on a dataset of individuals with movement disorders. Notably, our approach surpasses state-of-the-art methods in various scenarios, using fewer resources and proving highly suitable for clinical application, particularly in resource-constrained environments., Comment: Accepted for presentation at The IEEE International Symposium on Biomedical Imaging (ISBI 2025)

Biomedical Knowledge Graphs (BKGs) integrate diverse datasets to elucidate complex relationships within the biomedical field. Effective link prediction on these graphs can uncover valuable connections, such as potential novel drug-disease relations. We introduce a novel multimodal approach that unifies embeddings from specialized Language Models (LMs) with Graph Contrastive Learning (GCL) to enhance intra-entity relationships while employing a Knowledge Graph Embedding (KGE) model to capture inter-entity relationships for effective link prediction. To address limitations in existing BKGs, we present PrimeKG++, an enriched knowledge graph incorporating multimodal data, including biological sequences and textual descriptions for each entity type. By combining semantic and relational information in a unified representation, our approach demonstrates strong generalizability, enabling accurate link predictions even for unseen nodes. Experimental results on PrimeKG++ and the DrugBank drug-target interaction dataset demonstrate the effectiveness and robustness of our method across diverse biomedical datasets. Our source code, pre-trained models, and data are publicly available at https://github.com/HySonLab/BioMedKG

Prostate cancer, a growing global health concern, necessitates precise diagnostic tools, with Magnetic Resonance Imaging (MRI) offering high-resolution soft tissue imaging that significantly enhances diagnostic accuracy. Recent advancements in explainable AI and representation learning have significantly improved prostate cancer diagnosis by enabling automated and precise lesion classification. However, existing explainable AI methods, particularly those based on frameworks like generative adversarial networks (GANs), are predominantly developed for natural image generation, and their application to medical imaging often leads to suboptimal performance due to the unique characteristics and complexity of medical image. To address these challenges, our paper introduces three key contributions. First, we propose ProjectedEx, a generative framework that provides interpretable, multi-attribute explanations, effectively linking medical image features to classifier decisions. Second, we enhance the encoder module by incorporating feature pyramids, which enables multiscale feedback to refine the latent space and improves the quality of generated explanations. Additionally, we conduct comprehensive experiments on both the generator and classifier, demonstrating the clinical relevance and effectiveness of ProjectedEx in enhancing interpretability and supporting the adoption of AI in medical settings. Code will be released at https://github.com/Richardqiyi/ProjectedEx

The goal of voice conversion is to transform the speech of a source speaker to sound like that of a reference speaker while preserving the original content. A key challenge is to extract disentangled linguistic content from the source and voice style from the reference. While existing approaches leverage various methods to isolate the two, a generalization still requires further attention, especially for robustness in zero-shot scenarios. In this paper, we achieve successful disentanglement of content and speaker features by tuning self-supervised speech features with adapters. The adapters are trained to dynamically encode nuanced features from rich self-supervised features, and the decoder fuses them to produce speech that accurately resembles the reference with minimal loss of content. Moreover, we leverage a conditional flow matching decoder with cross-attention speaker conditioning to further boost the synthesis quality and efficiency. Subjective and objective evaluations in a zero-shot scenario demonstrate that the proposed method outperforms existing models in speech quality and similarity to the reference speech., Comment: ICASSP 2025; demo available https://mm.kaist.ac.kr/projects/AdaptVC

Dataset deduplication plays a crucial role in enhancing data quality, ultimately improving training performance and efficiency of LLMs. A commonly used method for data deduplication is the MinHash LSH algorithm. Recently, NVIDIA introduced a GPU-based MinHash LSH deduplication method, but it remains suboptimal, leaving room for further improvement in processing efficiency. This paper proposes a GPU-accelerated deduplication framework \sys that optimizes MinHash LSH for GPU clusters and leverages computationally efficient and partially reusable non-cryptographic hash functions. \sys significantly outperforms the CPU-based deduplication tool included in SlimPajama by up to 58.3 times and the GPU-based deduplication tool included in NVIDIA NeMo Curator by up to 8.6 times when processing 1 million documents with a node of four GPUs. Deduplication of 1.2 trillion tokens is completed in just 5.1 hours in a four-node, 16-GPU environment. The related code is publicly available on GitHub (https://github.com/mcrl/FED)., Comment: 12 pages, 3 figures

The Hierarchical Directed Capacitated Arc Routing Problem (HDCARP) is an extension of the Capacitated Arc Routing Problem (CARP), where the arcs of a graph are divided into classes based on their priority. The traversal of these classes is determined by either precedence constraints or a hierarchical objective, resulting in two distinct HDCARP variants. To the best of our knowledge, only one matheuristic has been proposed for these variants, but it performs relatively slowly, particularly for large-scale instances (Ha et al., 2024). In this paper, we propose a fast heuristic to efficiently address the computational challenges of HDCARP. Furthermore, we incorporate Reinforcement Learning (RL) into our heuristic to effectively guide the selection of local search operators, resulting in a hybrid algorithm. We name this hybrid algorithm as the Hybrid Reinforcement Learning and Heuristic Algorithm for Directed Arc Routing (HRDA). The hybrid algorithm adapts to changes in the problem dynamically, using real-time feedback to improve routing strategies and solution's quality by integrating heuristic methods. Extensive computational experiments on artificial instances demonstrate that this hybrid approach significantly improves the speed of the heuristic without deteriorating the solution quality. Our source code is publicly available at: https://github.com/HySonLab/ArcRoute

Clinician notes are a rich source of patient information but often contain inconsistencies due to varied writing styles, colloquialisms, abbreviations, medical jargon, grammatical errors, and non-standard formatting. These inconsistencies hinder the extraction of meaningful data from electronic health records (EHRs), posing challenges for quality improvement, population health, precision medicine, decision support, and research. We present a large language model approach to standardizing a corpus of 1,618 clinical notes. Standardization corrected an average of $4.9 +/- 1.8$ grammatical errors, $3.3 +/- 5.2$ spelling errors, converted $3.1 +/- 3.0$ non-standard terms to standard terminology, and expanded $15.8 +/- 9.1$ abbreviations and acronyms per note. Additionally, notes were re-organized into canonical sections with standardized headings. This process prepared notes for key concept extraction, mapping to medical ontologies, and conversion to interoperable data formats such as FHIR. Expert review of randomly sampled notes found no significant data loss after standardization. This proof-of-concept study demonstrates that standardization of clinical notes can improve their readability, consistency, and usability, while also facilitating their conversion into interoperable data formats.

We revisit the definition of the leading-twist chiral-even generalized parton distributions (GPDs) for $N \to \Delta$ baryon transitions. We identify and address deficiencies in previous definitions of the transition GPDs inspired by the transition form factors of the vector and axial-vector currents. Through systematic analysis of all possible covariant structures, respecting discrete symmetries and the baryon spinor equations of motion, we derive complete sets of independent structures for the transition matrix elements of the vector and axial-vector partonic operators. They contain additional structures proportional to the light-cone vector, corresponding to transition GPDs of vanishing first moment, which were not included in previous parametrizations. Their presence is confirmed independently by the light-front multipole expansion and the cross-channel SO(3) partial-wave analysis of the transition matrix elements. Our analysis provides a complete definition of the $N \to \Delta$ transition GPDs for use in theoretical and phenomenological studies., Comment: 13 pages

Graph Neural Networks (GNNs) have emerged as a critical tool for optimizing and managing the complexities of the Internet of Things (IoT) in next-generation networks. This survey presents a comprehensive exploration of how GNNs may be harnessed in 6G IoT environments, focusing on key challenges and opportunities through a series of open questions. We commence with an exploration of GNN paradigms and the roles of node, edge, and graph-level tasks in solving wireless networking problems and highlight GNNs' ability to overcome the limitations of traditional optimization methods. This guidance enhances problem-solving efficiency across various next-generation (NG) IoT scenarios. Next, we provide a detailed discussion of the application of GNN in advanced NG enabling technologies, including massive MIMO, reconfigurable intelligent surfaces, satellites, THz, mobile edge computing (MEC), and ultra-reliable low latency communication (URLLC). We then delve into the challenges posed by adversarial attacks, offering insights into defense mechanisms to secure GNN-based NG-IoT networks. Next, we examine how GNNs can be integrated with future technologies like integrated sensing and communication (ISAC), satellite-air-ground-sea integrated networks (SAGSIN), and quantum computing. Our findings highlight the transformative potential of GNNs in improving efficiency, scalability, and security within NG-IoT systems, paving the way for future advances. Finally, we propose a set of design guidelines to facilitate the development of efficient, scalable, and secure GNN models tailored for NG IoT applications., Comment: 28 pages, 15 figures, and 6 tables. Submitted for publication

The goal of this work is to generate natural speech in multiple languages while maintaining the same speaker identity, a task known as cross-lingual speech synthesis. A key challenge of cross-lingual speech synthesis is the language-speaker entanglement problem, which causes the quality of cross-lingual systems to lag behind that of intra-lingual systems. In this paper, we propose CrossSpeech++, which effectively disentangles language and speaker information and significantly improves the quality of cross-lingual speech synthesis. To this end, we break the complex speech generation pipeline into two simple components: language-dependent and speaker-dependent generators. The language-dependent generator produces linguistic variations that are not biased by specific speaker attributes. The speaker-dependent generator models acoustic variations that characterize speaker identity. By handling each type of information in separate modules, our method can effectively disentangle language and speaker representation. We conduct extensive experiments using various metrics, and demonstrate that CrossSpeech++ achieves significant improvements in cross-lingual speech synthesis, outperforming existing methods by a large margin.

In this paper, we consider a class of the Caputo fractional stochastic differential equations of fractional order $\alpha \in (\frac{1}{2},1]$. Our aim is to provide a complete analysis of the continuous dependence of solutions on the fractional order $\alpha.$ We first provide optimal estimates for the strong and weak distances between two solutions with different fractional orders. We then describe the exact asymptotic behavior of these distances., Comment: 34 pages

We present VoiceDiT, a multi-modal generative model for producing environment-aware speech and audio from text and visual prompts. While aligning speech with text is crucial for intelligible speech, achieving this alignment in noisy conditions remains a significant and underexplored challenge in the field. To address this, we present a novel audio generation pipeline named VoiceDiT. This pipeline includes three key components: (1) the creation of a large-scale synthetic speech dataset for pre-training and a refined real-world speech dataset for fine-tuning, (2) the Dual-DiT, a model designed to efficiently preserve aligned speech information while accurately reflecting environmental conditions, and (3) a diffusion-based Image-to-Audio Translator that allows the model to bridge the gap between audio and image, facilitating the generation of environmental sound that aligns with the multi-modal prompts. Extensive experimental results demonstrate that VoiceDiT outperforms previous models on real-world datasets, showcasing significant improvements in both audio quality and modality integration., Comment: Accepted to ICASSP 2025

Deep learning-based object detection has become ubiquitous in the last decade due to its high accuracy in many real-world applications. With this growing trend, these models are interested in being attacked by adversaries, with most of the results being on classifiers, which do not match the context of practical object detection. In this work, we propose a novel method to fool object detectors, expose the vulnerability of state-of-the-art detectors, and promote later works to build more robust detectors to adversarial examples. Our method aims to generate adversarial images by perturbing object confidence scores during training, which is crucial in predicting confidence for each class in the testing phase. Herein, we provide a more intuitive technique to embed additive noises based on detected objects' masks and the training loss with distortion control over the original image by leveraging the gradient of iterative images. To verify the proposed method, we perform adversarial attacks against different object detectors, including the most recent state-of-the-art models like YOLOv8, Faster R-CNN, RetinaNet, and Swin Transformer. We also evaluate our technique on MS COCO 2017 and PASCAL VOC 2012 datasets and analyze the trade-off between success attack rate and image distortion. Our experiments show that the achievable success attack rate is up to $100$\% and up to $98$\% when performing white-box and black-box attacks, respectively. The source code and relevant documentation for this work are available at the following link: https://github.com/anonymous20210106/attack_detector

Recent experiments of photon-nucleon and meson-nucleon scatterings have accumulated a lot of data for various meson production processes. One of the purposes of those experiments is to search for the missing resonances which are not discovered until now but whose existence was predicted by hadron models. The analyses of the data requires the development of dynamical coupled-channel models. Since several missing resonances are expected to have spin higher than 3/2, we need to include higher-spin resonances in dynamical coupled-channel models, which enable us to determine the couplings of effective Lagrangians of higher-spin baryons with pseudoscalar mesons or vector mesons. However, hadron models, such as quark models, give predictions only of the decay amplitudes of such baryons. Here we demonstrate the formalism of high-spin resonances and construct the relation between the coupling constants of effective Lagrangians and the partial decay widths that can be predicted by hadron models. This allows us to compare the coupling constants to the hadron model predictions not only in magnitude but in sign as well., Comment: 20 pages, 2 figures

DC motors have been widely used in many industrial applications, from small jointed robots with multiple degrees of freedom to household appliances and transportation vehicles such as electric cars and trains. The main function of these motors is to ensure stable positioning performance and speed for mechanical systems based on pre-designed control methods. However, achieving optimal speed performance for servo motors faces many challenges due to the impact of internal and external loads, which affect output stability. To optimize the speed performance of DC Servo motors, a control method combining PID controllers and artificial neural networks has been proposed. Traditional PID controllers have the advantage of a simple structure and effective control capability in many systems, but they face difficulties when dealing with nonlinear and uncertain changes. The neural network is integrated to adjust the PID parameters in real time, helping the system adapt to different operating conditions. Simulation and experimental results have demonstrated that the proposed method significantly improves the speed tracking capability and stability of the motor while ensuring quick response, zero steady-state error, and eliminating overshoot. This method offers high potential for application in servo motor control systems requiring high precision and performance., Comment: in Vietnamese language. Hoi nghi Quoc gia ve Dien tu, Truyen thong va Cong nghe Thong tin lan thu XXVII, REV-ECIT 2024

Hard combinatorial optimization problems, often mapped to Ising models, promise potential solutions with quantum advantage but are constrained by limited qubit counts in near-term devices. We present an innovative quantum-inspired framework that dynamically compresses large Ising models to fit available quantum hardware of different sizes. Thus, we aim to bridge the gap between large-scale optimization and current hardware capabilities. Our method leverages a physics-inspired GNN architecture to capture complex interactions in Ising models and accurately predict alignments among neighboring spins (aka qubits) at ground states. By progressively merging such aligned spins, we can reduce the model size while preserving the underlying optimization structure. It also provides a natural trade-off between the solution quality and size reduction, meeting different hardware constraints of quantum computing devices. Extensive numerical studies on Ising instances of diverse topologies show that our method can reduce instance size at multiple levels with virtually no losses in solution quality on the latest D-wave quantum annealers., Comment: Accepted to AAAI'25

Bundle recommendation aims to suggest a set of interconnected items to users. However, diverse interaction types and sparse interaction matrices often pose challenges for previous approaches in accurately predicting user-bundle adoptions. Inspired by the distant supervision strategy and generative paradigm, we propose BRIDGE, a novel framework for bundle recommendation. It consists of two main components namely the correlation-based item clustering and the pseudo bundle generation modules. Inspired by the distant supervision approach, the former is to generate more auxiliary information, e.g., instructive item clusters, for training without using external data. This information is subsequently aggregated with collaborative signals from user historical interactions to create pseudo `ideal' bundles. This capability allows BRIDGE to explore all aspects of bundles, rather than being limited to existing real-world bundles. It effectively bridging the gap between user imagination and predefined bundles, hence improving the bundle recommendation performance. Experimental results validate the superiority of our models over state-of-the-art ranking-based methods across five benchmark datasets.

We explore the production of gravitational waves (GW) resulting from a first-order phase transition (FOPT) in a non-minimally coupled `Dark Higgs Inflation' model. Utilizing a dark sector scalar field as the inflaton, we demonstrate how inflationary dynamics naturally set the stage for observable FOPT. These transitions, influenced by thermal and quantum effects, generate GW spectra potentially detectable by observatories such as LISA, DECIGO, the Cosmic Explorer and the Einstein Telescope. Our study highlights the inflaton's dual role in cosmic inflation and early Universe phase transitions, presenting a unified framework to probe physics beyond the Standard Model through gravitational wave astronomy., Comment: 24 pages, 9 figures

Recent probabilistic methods for 3D triangular meshes capture diverse shapes by differentiable mesh connectivity, but face high computational costs with increased shape details. We introduce a new differentiable mesh processing method in 2D and 3D that addresses this challenge and efficiently handles meshes with intricate structures. Additionally, we present an algorithm that adapts the mesh resolution to local geometry in 2D for efficient representation. We demonstrate the effectiveness of our approach on 2D point cloud and 3D multi-view reconstruction tasks. Visit our project page (https://sonsang.github.io/dmesh2-project) for source code and supplementary material., Comment: 26 pages, 27 figures, 4 tables

Visual Language Models have demonstrated remarkable capabilities across tasks, including visual question answering and image captioning. However, most models rely on text-based instructions, limiting their effectiveness in human-machine interactions. Moreover, the quality of language models depends on reasoning and prompting techniques, such as COT, which remain underexplored when using speech instructions. To address these challenges, we propose SilVar, a novel end-to-end multimodal model that uses speech instructions for reasoning in visual question answering. In addition, we investigate reasoning techniques with levels including conversational, simple, and complex speech instruction. SilVar is built upon CLIP, Whisper, and LLaMA 3.1-8B, enabling intuitive interactions by allowing users to provide verbal or text instructions. To this end, we introduce a dataset designed to challenge models with speech-based reasoning tasks for object localization. This dataset enhances the model ability to process and explain visual scenes from spoken input, moving beyond object recognition to reasoning-based interactions. The experiments show that SilVar achieves SOTA performance on the MMMU and ScienceQA benchmarks despite the challenge of speech-based instructions. We believe SilVar will inspire next-generation multimodal reasoning models, toward expert artificial general intelligence. Our code and dataset are available here., Comment: 10 pages

We present a parallelized differentiable traffic simulator based on the Intelligent Driver Model (IDM), a car-following framework that incorporates driver behavior as key variables. Our simulator efficiently models vehicle motion, generating trajectories that can be supervised to fit real-world data. By leveraging its differentiable nature, IDM parameters are optimized using gradient-based methods. With the capability to simulate up to 2 million vehicles in real time, the system is scalable for large-scale trajectory optimization. We show that we can use the simulator to filter noise in the input trajectories (trajectory filtering), reconstruct dense trajectories from sparse ones (trajectory reconstruction), and predict future trajectories (trajectory prediction), with all generated trajectories adhering to physical laws. We validate our simulator and algorithm on several datasets including NGSIM and Waymo Open Dataset., Comment: 8 pages, 6 figures, 2 tables

We introduce a new predator-prey model by replacing the growth and predation constant by a square matrix, and the population density as a population vector. The classical Lotka-Volterra model describes a population that either modulates or converges. Stability analysis of such models have been extensively studied by the works of Merdan (https://doi.org/10.1016/j.chaos.2007.06.062). The new model adds complexity by introducing an age group structure where the population of each age group evolves as prescribed by the Leslie matrix. The added complexity changes the behavior of the model such that the population either displays roughly an exponential growth or decay. We first provide an exact equation that describes a time evolution and use analytic techniques to obtain an approximate growth factor. We also discuss the variants of the Leslie model, i.e., the complex value predator-prey model and the competitive model. We then prove the Last Species Standing theorem that determines the dominant population in the large time limit. The recursive structure of the model denies the application of simple regression. We discuss a machine learning scheme that allows an admissible fit for the population evolution of Paramecium Aurelia and Paramecium Caudatum. Another potential avenue to simplify the computation is to use the machinery of quantum operators. We demonstrate the potential of this approach by computing the Hamiltonian of a simple Leslie system.

Emotion Recognition in Conversations (ERC) facilitates a deeper understanding of the emotions conveyed by speakers in each utterance within a conversation. Recently, Graph Neural Networks (GNNs) have demonstrated their strengths in capturing data relationships, particularly in contextual information modeling and multimodal fusion. However, existing methods often struggle to fully capture the complex interactions between multiple modalities and conversational context, limiting their expressiveness. To overcome these limitations, we propose ConxGNN, a novel GNN-based framework designed to capture contextual information in conversations. ConxGNN features two key parallel modules: a multi-scale heterogeneous graph that captures the diverse effects of utterances on emotional changes, and a hypergraph that models the multivariate relationships among modalities and utterances. The outputs from these modules are integrated into a fusion layer, where a cross-modal attention mechanism is applied to produce a contextually enriched representation. Additionally, ConxGNN tackles the challenge of recognizing minority or semantically similar emotion classes by incorporating a re-weighting scheme into the loss functions. Experimental results on the IEMOCAP and MELD benchmark datasets demonstrate the effectiveness of our method, achieving state-of-the-art performance compared to previous baselines.

The Mekong Delta Region of Vietnam faces increasing dengue risks driven by urbanization, globalization, and climate change. This study introduces a probabilistic forecasting model for predicting dengue incidence and outbreaks with one to three month lead times, integrating meteorological, sociodemographic, preventive, and epidemiological data. Seventy-two models were evaluated, and an ensemble combining top-performing spatiotemporal, supervised PCA, and semi-mechanistic hhh4 frameworks was developed. Using data from 2004-2022 for training, validation, and evaluation, the ensemble model demonstrated 69% accuracy at a 3-month horizon, outperforming a baseline model. While effective, its performance declined in years with atypical seasonality, such as 2019 and 2022. The model provides critical lead time for targeted dengue prevention and control measures, addressing a growing public health need in the region., Comment: 34 pages, 6 figures

When trained with severely imbalanced data, deep neural networks often struggle to accurately recognize classes with only a few samples. Previous studies in long-tailed recognition have attempted to rebalance biased learning using known sample distributions, primarily addressing different classification difficulties at the class level. However, these approaches often overlook the instance difficulty variation within each class. In this paper, we propose a difficulty-aware balancing margin (DBM) loss, which considers both class imbalance and instance difficulty. DBM loss comprises two components: a class-wise margin to mitigate learning bias caused by imbalanced class frequencies, and an instance-wise margin assigned to hard positive samples based on their individual difficulty. DBM loss improves class discriminativity by assigning larger margins to more difficult samples. Our method seamlessly combines with existing approaches and consistently improves performance across various long-tailed recognition benchmarks.

We propose a new view synthesis method via synthesizing a 3D neural field from both single or few-view input images. To address the ill-posed nature of the image-to-3D generation problem, we devise a two-stage method that involves a reconstruction model and a diffusion model for view synthesis. Our reconstruction model first lifts one or more input images to the 3D space from a volume as the coarse-scale 3D representation followed by a tri-plane as the fine-scale 3D representation. To mitigate the ambiguity in occluded regions, our diffusion model then hallucinates missing details in the rendered images from tri-planes. We then introduce a new progressive refinement technique that iteratively applies the reconstruction and diffusion model to gradually synthesize novel views, boosting the overall quality of the 3D representations and their rendering. Empirical evaluation demonstrates the superiority of our method over state-of-the-art methods on the synthetic SRN-Car dataset, the in-the-wild CO3D dataset, and large-scale Objaverse dataset while achieving both sampling efficacy and multi-view consistency.

Video Action Detection (VAD) involves localizing and categorizing action instances in videos. Videos inherently contain various information sources, including audio, visual cues, and surrounding scene contexts. Effectively leveraging this multi-modal information for VAD is challenging, as the model must accurately focus on action-relevant cues. In this study, we introduce a novel multi-modal VAD architecture called the Joint Actor-centric Visual, Audio, Language Encoder (JoVALE). JoVALE is the first VAD method to integrate audio and visual features with scene descriptive context derived from large image captioning models. The core principle of JoVALE is the actor-centric aggregation of audio, visual, and scene descriptive contexts, where action-related cues from each modality are identified and adaptively combined. We propose a specialized module called the Actor-centric Multi-modal Fusion Network, designed to capture the joint interactions among actors and multi-modal contexts through Transformer architecture. Our evaluation conducted on three popular VAD benchmarks, AVA, UCF101-24, and JHMDB51-21, demonstrates that incorporating multi-modal information leads to significant performance gains. JoVALE achieves state-of-the-art performances. The code will be available at \texttt{https://github.com/taeiin/AAAI2025-JoVALE}., Comment: Accepted to AAAI Conference on Artificial Intelligence 2025, 9 pages, 5 figures

Image to image matching has been well studied in the computer vision community. Previous studies mainly focus on training a deep metric learning model matching visual patterns between the query image and gallery images. In this study, we show that pure image-to-image matching suffers from false positives caused by matching to local visual patterns. To alleviate this issue, we propose to leverage recent advances in vision-language pretraining research. Specifically, we introduce additional image-text alignment losses into deep metric learning, which serve as constraints to the image-to-image matching loss. With additional alignments between the text (e.g., product title) and image pairs, the model can learn concepts from both modalities explicitly, which avoids matching low-level visual features. We progressively develop two variants, a 3-tower and a 4-tower model, where the latter takes one more short text query input. Through extensive experiments, we show that this change leads to a substantial improvement to the image to image matching problem. We further leveraged this model for multimodal search, which takes both image and reformulation text queries to improve search quality. Both offline and online experiments show strong improvements on the main metrics. Specifically, we see 4.95% relative improvement on image matching click through rate with the 3-tower model and 1.13% further improvement from the 4-tower model.

Visual-Language Models (VLMs) have become a powerful tool for bridging the gap between visual and linguistic understanding. However, the conventional learning approaches for VLMs often suffer from limitations, such as the high resource requirements of collecting and training image-text paired data. Recent research has suggested that language understanding plays a crucial role in the performance of VLMs, potentially indicating that text-only training could be a viable approach. In this work, we investigate the feasibility of enhancing fine-grained visual understanding in VLMs through text-only training. Inspired by how humans develop visual concept understanding, where rich textual descriptions can guide visual recognition, we hypothesize that VLMs can also benefit from leveraging text-based representations to improve their visual recognition abilities. We conduct comprehensive experiments on two distinct domains: fine-grained species classification and cultural visual understanding tasks. Our findings demonstrate that text-only training can be comparable to conventional image-text training while significantly reducing computational costs. This suggests a more efficient and cost-effective pathway for advancing VLM capabilities, particularly valuable in resource-constrained environments., Comment: AAAI25 workshop accepted

We explored in this work the ubiquitous phenomenon of serial scammers, who deploy thousands of addresses to conduct a series of similar Rug Pulls on popular decentralized exchanges (DEXs). We first constructed a list of about 384,000 scammer addresses behind all 1-day Rug Pulls on the two most popular DEXs, Uniswap (Ethereum) and Pancakeswap (BSC), and identified many distinctive scam patterns including star-shaped, chain-shaped, and majority-flow scam clusters. We then proposed an algorithm to build a complete scam network from given scammer addresses, which consists of not only scammer addresses but also supporting addresses including depositors, withdrawers, transferrers, coordinators, and most importantly, wash traders. We note that profit estimations in existing works on Rug Pulls failed to capture the cost of wash trading, leading to inflated figures. Knowing who the wash traders are, we established a more accurate estimate for the true profit of individual scam pools as well as of the entire (serial) scam network by taking into account the wash-trading expenses.

Despite remarkable progress in knowledge transfer across visual and textual domains, extending these achievements to indoor localization, particularly for learning transferable representations among Received Signal Strength (RSS) fingerprint datasets, remains a challenge. This is due to inherent discrepancies among these RSS datasets, largely including variations in building structure, the input number and disposition of WiFi anchors. Accordingly, specialized networks, which were deprived of the ability to discern transferable representations, readily incorporate environment-sensitive clues into the learning process, hence limiting their potential when applied to specific RSS datasets. In this work, we propose a plug-and-play (PnP) framework of knowledge transfer, facilitating the exploitation of transferable representations for specialized networks directly on target RSS datasets through two main phases. Initially, we design an Expert Training phase, which features multiple surrogate generative teachers, all serving as a global adapter that homogenizes the input disparities among independent source RSS datasets while preserving their unique characteristics. In a subsequent Expert Distilling phase, we continue introducing a triplet of underlying constraints that requires minimizing the differences in essential knowledge between the specialized network and surrogate teachers through refining its representation learning on the target dataset. This process implicitly fosters a representational alignment in such a way that is less sensitive to specific environmental dynamics. Extensive experiments conducted on three benchmark WiFi RSS fingerprint datasets underscore the effectiveness of the framework that significantly exerts the full potential of specialized networks in localization., Comment: Accepted in the 1st round at WACV 2025 (Algorithm Track)

We propose iMoT, an innovative Transformer-based inertial odometry method that retrieves cross-modal information from motion and rotation modalities for accurate positional estimation. Unlike prior work, during the encoding of the motion context, we introduce Progressive Series Decoupler at the beginning of each encoder layer to stand out critical motion events inherent in acceleration and angular velocity signals. To better aggregate cross-modal interactions, we present Adaptive Positional Encoding, which dynamically modifies positional embeddings for temporal discrepancies between different modalities. During decoding, we introduce a small set of learnable query motion particles as priors to model motion uncertainties within velocity segments. Each query motion particle is intended to draw cross-modal features dedicated to a specific motion mode, all taken together allowing the model to refine its understanding of motion dynamics effectively. Lastly, we design a dynamic scoring mechanism to stabilize iMoT's optimization by considering all aligned motion particles at the final decoding step, ensuring robust and accurate velocity segment estimation. Extensive evaluations on various inertial datasets demonstrate that iMoT significantly outperforms state-of-the-art methods in delivering superior robustness and accuracy in trajectory reconstruction., Comment: Accepted as technical research paper in 39th AAAI Conference on Artificial Intelligence, 2025 (AAAI 2025)

We propose a nested quantum dot structure for improved control of entanglement induced by the Heisenberg exchange between an electron and a qubit with relative motion. The entanglement is quantified by the mutual information (MI). The electron, initially prepared in the ground state, generally produces greater entanglement when excited to the scattering state compared to remaining in the bound state. In the bound state, the final entanglement oscillates as a function of the qubit speed and can be tuned accordingly. In the case of long-range interaction, the normalized exchange distribution leads to substantial final entanglement, independent of the qubit moving direction, indicating that even very weak but prolonged exchange can still generate significant entanglement. In the case of short-range interaction, different moving directions lead to varying MI values. We also consider the scenario without the nested dot and find that the same maximum (among all times) MI is pre-determined solely by the initial angle between the spins. In this case, the entanglement exhibits different growth characteristics during different phases. The saturation of the MI mimics that of a strict zero-dimensional quantum dot, where exchange and time are combined into a single parameter, the amount of interaction., Comment: 7 pages, 3 figures

Most techniques approach the problem of image forgery localization as a binary segmentation task, training neural networks to label original areas as 0 and forged areas as 1. In contrast, we tackle this issue from a more fundamental perspective by partitioning images according to their originating sources. To this end, we propose Segment Any Forged Image Region (SAFIRE), which solves forgery localization using point prompting. Each point on an image is used to segment the source region containing itself. This allows us to partition images into multiple source regions, a capability achieved for the first time. Additionally, rather than memorizing certain forgery traces, SAFIRE naturally focuses on uniform characteristics within each source region. This approach leads to more stable and effective learning, achieving superior performance in both the new task and the traditional binary forgery localization., Comment: Accepted at AAAI 2025. Code is available at: https://github.com/mjkwon2021/SAFIRE

Given a simple graph $G$, the artinian monomial algebra associated to $G$, denoted by $A(G)$, is defined by the edge ideal of $G$ and the squares of the variables. In this article, we classify some tadpole graphs $G$ for which $A(G)$ has or fails the weak Lefschetz property., Comment: 12 pages. arXiv admin note: substantial text overlap with arXiv:2310.14368 by other authors

Large language models (LLMs) often struggle to objectively identify latent characteristics in large datasets due to their reliance on pre-trained knowledge rather than actual data patterns. To address this data grounding issue, we propose Data Scientist AI (DSAI), a framework that enables unbiased and interpretable feature extraction through a multi-stage pipeline with quantifiable prominence metrics for evaluating extracted features. On synthetic datasets with known ground-truth features, DSAI demonstrates high recall in identifying expert-defined features while faithfully reflecting the underlying data. Applications on real-world datasets illustrate the framework's practical utility in uncovering meaningful patterns with minimal expert oversight, supporting use cases such as interpretable classification. The title of our paper is chosen from multiple candidates based on DSAI-generated criteria.

In resource theories, catalysis refers to the possibility of enabling otherwise inaccessible quantum state transitions by providing the agent with an auxiliary system, under the condition that this auxiliary is returned to its initial state at the end of the protocol. Most studies to date have focused on fine-tuned catalytic processes that are highly sensitive to error: if the initial state of the system deviates even slightly from that for which the catalyst was designed, the catalyst would be irreparably degraded. To address this challenge, we introduce and study robust catalytic transformations and explore the extent of their capabilities. It turns out that robust catalysis is subtly related to the property of resource broadcasting. In particular, we show that the possibility of robust catalysis is equivalent to that of resource broadcasting in completely resource non-generating theories. This allows us to characterize a general class of resource theories that allow neither robust catalysis nor resource broadcasting, and another class where instead resource broadcasting and robust catalysis are possible and provide maximal advantage. Our approach encompasses a wide range of quantum resource theories, including entanglement, coherence, thermodynamics, magic, and imaginarity., Comment: 14 pages, 2 figures, and 1 table

Efficiently preparing approximate ground-states of large, strongly correlated systems on quantum hardware is challenging and yet nature is innately adept at this. This has motivated the study of thermodynamically inspired approaches to ground-state preparation that aim to replicate cooling processes via imaginary-time evolution. However, synthesizing quantum circuits that efficiently implement imaginary-time evolution is itself difficult, with prior proposals generally adopting heuristic variational approaches or using deep block encodings. Here, we use the insight that quantum imaginary-time evolution is a solution of Brockett's double-bracket flow and synthesize circuits that implement double-bracket flows coherently on the quantum computer. We prove that our Double-Bracket Quantum Imaginary-Time Evolution (DB-QITE) algorithm inherits the cooling guarantees of imaginary-time evolution. Concretely, each step is guaranteed to i) decrease the energy of an initial approximate ground-state by an amount proportion to the energy fluctuations of the initial state and ii) increase the fidelity with the ground-state. Thus DB-QITE provides a means to systematically improve the approximation of a ground-state using shallow circuits.

Inverse problems involving partial differential equations (PDEs) can be seen as discovering a mapping from measurement data to unknown quantities, often framed within an operator learning approach. However, existing methods typically rely on large amounts of labeled training data, which is impractical for most real-world applications. Moreover, these supervised models may fail to capture the underlying physical principles accurately. To address these limitations, we propose a novel architecture called Physics-Informed Deep Inverse Operator Networks (PI-DIONs), which can learn the solution operator of PDE-based inverse problems without labeled training data. We extend the stability estimates established in the inverse problem literature to the operator learning framework, thereby providing a robust theoretical foundation for our method. These estimates guarantee that the proposed model, trained on a finite sample and grid, generalizes effectively across the entire domain and function space. Extensive experiments are conducted to demonstrate that PI-DIONs can effectively and accurately learn the solution operators of the inverse problems without the need for labeled data.