arXiv subjects

Nana Wang

Publications and source records attributed to Nana Wang.

At least 19 recordsLinked to original sources

arXiv2026

2K Retrofit: Entropy-Guided Efficient Sparse Refinement for High-Resolution 3D Geometry Prediction

High-resolution geometric prediction is essential for robust perception in autonomous driving, robotics, and AR/MR, but current foundation models are fundamentally limited by their scalability to real-world, high-resolution scenarios. Direct inference on 2K images with these models incurs prohibitive computational and memory demands, making practical deployment challenging. To tackle the issue, we present 2K Retrofit, a novel framework that enables efficient 2K-resolution inference for any geometric foundation model, without modifying or retraining the backbone. Our approach leverages fast coarse predictions and an entropy-based sparse refinement to selectively enhance high-uncertainty regions, achieving precise and high-fidelity 2K outputs with minimal overhead. Extensive experiments on widely used benchmark demonstrate that 2K Retrofit consistently achieves state-of-the-art accuracy and speed, bridging the gap between research advances and scalable deployment in high-resolution 3D vision applications. Code will be released upon acceptance.

cs.CV↗

arXiv2026

Topology Estimation for Open Multi-Agent Systems

We address the problem of interaction topology identification in open multi-agent systems (OMAS) with dynamic node sets and fast switching interactions. In such systems, new agents join and interactions change rapidly, resulting in intervals with short dwell time and rendering conventional segment-wise estimation and clustering methods unreliable. To overcome this, we propose a projection-based dissimilarity measure derived from a consistency property of local least-squares operators, enabling robust mode clustering. Aggregating intervals within each cluster yields accurate topology estimates. The proposed framework offers a systematic solution for reconstructing the interaction topology of OMAS subject to fast switching. Finally, we illustrate our theoretical results via numerical simulations.

eess.SY↗

arXiv2026

Topology Identification of Dynamical Signed Graphs

We propose an adaptive control protocol for identifying the topology of dynamical networks interconnected over undirected graphs with cooperative and antagonistic interactions. The signed network is modeled using a repelling Laplacian. Topology identification relies on an edge-based formulation of the network and adaptive control protocols through the design of a persistently excited auxiliary network. Our approach guarantees the simultaneous identification and synchronization of the unknown signed network and establishes uniform semiglobal practical asymptotic stability of the estimation errors. Numerical simulations validate our theoretical results.

eess.SY↗

arXiv2025

New Synthetic Goldmine: Hand Joint Angle-Driven EMG Data Generation Framework for Micro-Gesture Recognition

Electromyography (EMG)-based gesture recognition has emerged as a promising approach for human-computer interaction. However, its performance is often limited by the scarcity of labeled EMG data, significant cross-user variability, and poor generalization to unseen gestures. To address these challenges, we propose SeqEMG-GAN, a conditional, sequence-driven generative framework that synthesizes high-fidelity EMG signals from hand joint angle sequences. Our method introduces a context-aware architecture composed of an angle encoder, a dual-layer context encoder featuring the novel Ang2Gist unit, a deep convolutional EMG generator, and a discriminator, all jointly optimized via adversarial learning. By conditioning on joint kinematic trajectories, SeqEMG-GAN is capable of generating semantically consistent EMG sequences, even for previously unseen gestures, thereby enhancing data diversity and physiological plausibility. Experimental results show that classifiers trained solely on synthetic data experience only a slight accuracy drop (from 57.77\% to 55.71\%). In contrast, training with a combination of real and synthetic data significantly improves accuracy to 60.53\%, outperforming real-only training by 2.76\%. These findings demonstrate the effectiveness of our framework,also achieves the state-of-art performance in augmenting EMG datasets and enhancing gesture recognition performance for applications such as neural robotic hand control, AI/AR glasses, and gesture-based virtual gaming systems.

cs.HC↗

arXiv2025

Language-Driven Coordination and Learning in Multi-Agent Simulation Environments

This paper introduces LLM-MARL, a unified framework that incorporates large language models (LLMs) into multi-agent reinforcement learning (MARL) to enhance coordination, communication, and generalization in simulated game environments. The framework features three modular components of Coordinator, Communicator, and Memory, which dynamically generate subgoals, facilitate symbolic inter-agent messaging, and support episodic recall. Training combines PPO with a language-conditioned loss and LLM query gating. LLM-MARL is evaluated in Google Research Football, MAgent Battle, and StarCraft II. Results show consistent improvements over MAPPO and QMIX in win rate, coordination score, and zero-shot generalization. Ablation studies demonstrate that subgoal generation and language-based messaging each contribute significantly to performance gains. Qualitative analysis reveals emergent behaviors such as role specialization and communication-driven tactics. By bridging language modeling and policy learning, this work contributes to the design of intelligent, cooperative agents in interactive simulations. It offers a path forward for leveraging LLMs in multi-agent systems used for training, games, and human-AI collaboration.

cs.AI↗

arXiv2025

EMG-UP: Unsupervised Personalization in Cross-User EMG Gesture Recognition

Cross-user electromyography (EMG)-based gesture recognition represents a fundamental challenge in achieving scalable and personalized human-machine interaction within real-world applications. Despite extensive efforts, existing methodologies struggle to generalize effectively across users due to the intrinsic biological variability of EMG signals, resulting from anatomical heterogeneity and diverse task execution styles. To address this limitation, we introduce EMG-UP, a novel and effective framework for Unsupervised Personalization in cross-user gesture recognition. The proposed framework leverages a two-stage adaptation strategy: (1) Sequence-Cross Perspective Contrastive Learning, designed to disentangle robust and user-specific feature representations by capturing intrinsic signal patterns invariant to inter-user variability, and (2) Pseudo-Label-Guided Fine-Tuning, which enables model refinement for individual users without necessitating access to source domain data. Extensive evaluations show that EMG-UP achieves state-of-the-art performance, outperforming prior methods by at least 2.0% in accuracy.

cs.HC↗

arXiv2025

Undress to Redress: A Training-Free Framework for Virtual Try-On

Virtual try-on (VTON) is a crucial task for enhancing user experience in online shopping by generating realistic garment previews on personal photos. Although existing methods have achieved impressive results, they struggle with long-sleeve-to-short-sleeve conversions-a common and practical scenario-often producing unrealistic outputs when exposed skin is underrepresented in the original image. We argue that this challenge arises from the ''majority'' completion rule in current VTON models, which leads to inaccurate skin restoration in such cases. To address this, we propose UR-VTON (Undress-Redress Virtual Try-ON), a novel, training-free framework that can be seamlessly integrated with any existing VTON method. UR-VTON introduces an ''undress-to-redress'' mechanism: it first reveals the user's torso by virtually ''undressing,'' then applies the target short-sleeve garment, effectively decomposing the conversion into two more manageable steps. Additionally, we incorporate Dynamic Classifier-Free Guidance scheduling to balance diversity and image quality during DDPM sampling, and employ Structural Refiner to enhance detail fidelity using high-frequency cues. Finally, we present LS-TON, a new benchmark for long-sleeve-to-short-sleeve try-on. Extensive experiments demonstrate that UR-VTON outperforms state-of-the-art methods in both detail preservation and image quality. Code will be released upon acceptance.

cs.CV↗

arXiv2024

Simultaneous Topology Estimation and Synchronization of Dynamical Networks with Time-varying Topology

We propose an adaptive control strategy for the simultaneous estimation of topology and synchronization in complex dynamical networks with unknown, time-varying topology. Our approach transforms the problem of time-varying topology estimation into a problem of estimating the time-varying weights of a complete graph, utilizing an edge-agreement framework. We introduce two auxiliary networks: one that satisfies the persistent excitation condition to facilitate topology estimation, while the other, a uniform-$δ$ persistently exciting network, ensures the boundedness of both weight estimation and synchronization errors, assuming bounded time-varying weights and their derivatives. A relevant numerical example shows the efficiency of our methods.

cs.MA↗

arXiv2024

Differentially Private Spatiotemporal Trajectory Synthesis with Retained Data Utility

Spatiotemporal trajectories collected from GPS-enabled devices are of vital importance to many applications, such as urban planning and traffic analysis. Due to the privacy leakage concerns, many privacy-preserving trajectory publishing methods have been proposed. However, most of them could not strike a good balance between privacy protection and good data utility. In this paper, we propose DP-STTS, a differentially private spatiotemporal trajectory synthesizer with high data utility, which employs a model composed of a start spatiotemporal cube distribution and a 1-order Markov process. Specially, DP-STTS firstly discretizes the raw spatiotemporal trajectories into neighboring cubes, such that the model size is limited and the model's tolerance for noise could be enhanced. Then, a Markov process is utilized for the next location point picking. After adding noise under differential privacy (DP) to the model, synthetic trajectories that preserve essential spatial and temporal characteristics of the real trajectories are generated from the noisy model. Experiments on one real-life dataset demonstrate that DP-STTS provides good data utility. Our code is available at https://github.com/Etherious72/DP-STTS.

cs.CR↗

arXiv2024

DPTraj-PM: Differentially Private Trajectory Synthesis Using Prefix Tree and Markov Process

The increasing use of GPS-enabled devices has generated a large amount of trajectory data. These data offer us vital insights to understand the movements of individuals and populations, benefiting a broad range of applications from transportation planning to epidemic modeling. However, improper release of trajectory data is increasing concerns on individual privacy. Previous attempts either lack strong privacy guarantees, or fail to preserve sufficient basic characteristics of the original data. In this paper, we propose DPTraj-PM, a method to synthesize trajectory dataset under the differential privacy (DP) framework while ensures high data utility. Based on the assumption that an individual's trajectory could be mainly determined by the initial trajectory segment (which depicts the starting point and the initial direction) and the next location point, DPTraj-PM discretizes the raw trajectories into neighboring cells, and models them by combining a prefix tree structure and an m-order Markov process. After adding noise to the model under differential privacy, DPTraj-PM generates a synthetic dataset from the noisy model to enable a wider spectrum of data mining and modeling tasks. The output traces crafted by DPTraj-PM not only preserves the patterns and variability in individuals' mobility behaviors, but also protects individual privacy. Experiments on two real-world datasets demonstrate that DPTraj-PM substantially outperforms the state-of-the-art techniques in terms of data utility. Our code is available at https://github.com/wnn5/DP-PrefixTreeMarkov.

cs.CR↗

arXiv2023

A dynamic combustion model for supersonic turbulent combustion

Supersonic combustion plays a vital role in various applications, including scramjets, dual-mode ramjets, and pulse detonation engines. However, the flame characteristics can vary significantly, depending on the application. To model supersonic combustion, large eddy simulation coupled with a partially stirred reactor (PaSR) is commonly used. This method assumes that reactions occur at turbulent fine flame structure in a computational cell, making it less effective for flames governed by not only turbulent mixing but also homogenous autoignition, which is common under supersonic flow. To address this limitation, this study proposes a novel dynamic combustion model to enable more versatile modeling of supersonic flames. The model utilizes a two-delta probability density function, which represents the sub-grid composition variation and models the filtered sub-grid reaction rate. The corresponding weights are dynamically modeled based on the level of cell composition inhomogeneity. The new dynamic model is tested in a supersonic combustion case study with a strut-cavity flame holder. Results demonstrate that the new dynamic model can properly recover the limits of supersonic flames that are primarily governed by homogenous autoignition and turbulent mixing.

physics.flu-dyn↗

arXiv2021

OpenClinicalAI: enabling AI to diagnose diseases in real-world clinical settings

This paper quantitatively reveals the state-of-the-art and state-of-the-practice AI systems only achieve acceptable performance on the stringent conditions that all categories of subjects are known, which we call closed clinical settings, but fail to work in real-world clinical settings. Compared to the diagnosis task in the closed setting, real-world clinical settings pose severe challenges, and we must treat them differently. We build a clinical AI benchmark named Clinical AIBench to set up real-world clinical settings to facilitate researches. We propose an open, dynamic machine learning framework and develop an AI system named OpenClinicalAI to diagnose diseases in real-world clinical settings. The first versions of Clinical AIBench and OpenClinicalAI target Alzheimer's disease. In the real-world clinical setting, OpenClinicalAI significantly outperforms the state-of-the-art AI system. In addition, OpenClinicalAI develops personalized diagnosis strategies to avoid unnecessary testing and seamlessly collaborates with clinicians. It is promising to be embedded in the current medical systems to improve medical services.

cs.AI↗

arXiv2019

LoadCNN: A Low Training Cost Deep Learning Model for Day-Ahead Individual Residential Load Forecasting

Accurate day-ahead individual residential load forecasting is of great importance to various applications of smart grid on day-ahead market. Deep learning, as a powerful machine learning technology, has shown great advantages and promising application in load forecasting tasks. However, deep learning is a computationally-hungry method, and requires high costs (e.g., time, energy and CO2 emission) to train a deep learning model, which aggravates the energy crisis and incurs a substantial burden to the environment. As a consequence, the deep learning methods are difficult to be popularized and applied in the real smart grid environment. In this paper, we propose a low training cost model based on convolutional neural network, namely LoadCNN, for next-day load forecasting of individual resident with reduced training cost. The experiments show that the training time of LoadCNN is only approximately 1/54 of the one of other state-of-the-art models, and energy consumption and CO2 emissions are only approximate 1/45 of those of other state-of-the-art models based on the same indicators. Meanwhile, the prediction accuracy of our model is equal to that of current state-of-the-art models, making LoadCNN the first load forecasting model simultaneously achieving high prediction accuracy and low training costs. LoadCNN is an efficient green model that is able to be quickly, cost-effectively and environmentally-friendly deployed in a realistic smart grid environment.

eess.SP↗

arXiv2019

EasiCS: the objective and fine-grained classification method of cervical spondylosis dysfunction

The precise diagnosis is of great significance in developing precise treatment plans to restore neck function and reduce the burden posed by the cervical spondylosis (CS). However, the current available neck function assessment method are subjective and coarse-grained. In this paper, based on the relationship among CS, cervical structure, cervical vertebra function, and surface electromyography (sEMG), we seek to develop a clustering algorithms on the sEMG data set collected from the clinical environment and implement the division. We proposed and developed the framework EasiCS, which consists of dimension reduction, clustering algorithm EasiSOM, spectral clustering algorithm EasiSC. The EasiCS outperform the commonly used seven algorithms overall.

cs.LG↗

arXiv2019

A new direction to promote the implementation of artificial intelligence in natural clinical settings

Artificial intelligence (AI) researchers claim that they have made great `achievements' in clinical realms. However, clinicians point out the so-called `achievements' have no ability to implement into natural clinical settings. The root cause for this huge gap is that many essential features of natural clinical tasks are overlooked by AI system developers without medical background. In this paper, we propose that the clinical benchmark suite is a novel and promising direction to capture the essential features of the real-world clinical tasks, hence qualifies itself for guiding the development of AI systems, promoting the implementation of AI in real-world clinical practice.

cs.AI↗

arXiv2018

EasiCSDeep: A deep learning model for Cervical Spondylosis Identification using surface electromyography signal

Cervical spondylosis (CS) is a common chronic disease that affects up to two-thirds of the population and poses a serious burden on individuals and society. The early identification has significant value in improving cure rate and reducing costs. However, the pathology is complex, and the mild symptoms increase the difficulty of the diagnosis, especially in the early stage. Besides, the time-consuming and costliness of hospital medical service reduces the attention to the CS identification. Thus, a convenient, low-cost intelligent CS identification method is imperious demanded. In this paper, we present an intelligent method based on the deep learning to identify CS, using the surface electromyography (sEMG) signal. Faced with the complex, high dimensionality and weak usability of the sEMG signal, we proposed and developed a multi-channel EasiCSDeep algorithm based on the convolutional neural network, which consists of the feature extraction, spatial relationship representation and classification algorithm. To the best of our knowledge, this EasiCSDeep is the first effort to employ the deep learning and the sEMG data to identify CS. Compared with previous state-of-the-art algorithm, our algorithm achieves a significant improvement.

cs.LG↗

arXiv2018

Clustering Residential Electricity Load Curves via Community Detection in Network

Performing analytic of household load curves (LCs) has significant value in predicting individual electricity consumption patterns, and hence facilitate developing demand-response strategy, and finally achieve energy efficiency improvement and emission reduction. LC clustering is a widely used analytic technology, which discovers electricity consumption patterns by grouping similar LCs into same sub-groups. However, previous clustering methods treat each LC in the data set as an individual time series, ignoring the inherent relationship among different LCs, restraining the performance of the clustering methods. What's more, due to the significant volatility and uncertainty of LCs, the previous LC clustering approaches inevitably result in either lager number of clusters or huge variances within a cluster, which is unacceptable for actual application needs. In this paper, we proposed an integrated approach to address this issue. First, we converted the LC clustering task into the community detection task in network. Second, we proposed a clustering approach incorporated with community detection to improve the performance of LC clustering, which includes network construction, community detection and typical load profile extraction. The method convert the data set into a network in which the inherent relationship among LCs is represented by the edges of the network. Third, we construct a multi-layer typical load profile directory to make the trade-off between variances within a cluster and the number of the clusters, enabling the researchers to assess the LCs and the customers in different layers according to practical application requirements. The experiments demonstrate that our integrated approach outperform the state-of-the-art methods.

cs.DS↗

arXiv2018

Neighborhood selection with application to social networks

The topic of this paper is modeling and analyzing dependence in stochastic social networks. Using a latent variable block model allows the analysis of dependence between blocks via the analysis of a latent graphical model. Our approach to the analysis of the graphical model then is based on the idea underlying the neighborhood selection scheme put forward by Meinshausen and Bühlmann (2006). However, because of the latent nature of our model, estimates have to be used in lieu of the unobserved variables. This leads to a novel analysis of graphical models under uncertainty, in the spirit of Rosenbaum et al. (2010), or Belloni et al. (2017). Lasso-based selectors, and a class of Dantzig-type selectors are studied.

math.ST↗