arXiv subjects

Zitong Wang

Publications and source records attributed to Zitong Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

PonderLM-2: Pretraining LLM with Latent Thoughts in Continuous Space

The remarkable success of Chain-of-Thought (CoT), which enhances performance by scaling generation steps at test-time, inspires us to ask: can we leverage a similar scaling of computational steps during pretraining to improve the generation of each individual token? To address this, we propose a novel pre-training methodology: Pretraining Language Models with Latent Thoughts (PonderLM-2). Our approach pretrains a language model (LM) to first generate an intermediate latent thought-the last hidden state of the current position-which is then used as input to predict the actual subsequent token. This additional computational step enables the LM to refine its prediction within unconstrained continuous space. Our experiments demonstrate that, at an identical inference cost, a LM that generates one additional latent thought per token outperforms a standard model with double the parameters. For instance, our PonderLM-2-Pythia-1.4B, pretrained on 300B tokens from the Pile, significantly surpasses the vanilla Pythia-2.8B trained on the same data on both language modeling and a range of general downstream tasks. Furthermore, increasing the number of latent thoughts generated before each actual token-forming a chain analogous to CoT-consistently improves the model's performance. The code is available at https://github.com/LUMIA-Group/PonderLM-2.

cs.CL↗

arXiv2026

Graph-Based Multi-Modal Light-weight Network for Adaptive Brain Tumor Segmentation

Multi-modal brain tumor segmentation remains challenging for practical deployment due to the high computational costs of mainstream models. In this work, we propose GMLN-BTS, a Graph-based Multi-modal interaction Lightweight Network for brain tumor segmentation. Our architecture achieves high-precision, resource-efficient segmentation through three key components. First, a Modality-Aware Adaptive Encoder (M2AE) facilitates efficient multi-scale semantic extraction. Second, a Graph-based Multi-Modal Collaborative Interaction Module (G2MCIM) leverages graph structures to model complementary cross-modal relationships. Finally, a Voxel Refinement UpSampling Module (VRUM) integrates linear interpolation with multi-scale transposed convolutions to suppress artifacts and preserve boundary details. Experimental results on BraTS 2017, 2019, and 2021 benchmarks demonstrate that GMLN-BTS achieves state-of-the-art performance among lightweight models. With only 4.58M parameters, our method reduces parameter count by 98% compared to mainstream 3D Transformers while significantly outperforming existing compact approaches.

eess.IV↗

arXiv2026

Joint Optimization of Latency and Accuracy for Split Federated Learning in User-Centric Cell-Free MIMO Networks

This paper proposes a user-centric split federated learning (UCSFL) framework for user-centric cell-free multiple-input multiple-output (CF-MIMO) networks to support split federated learning (SFL). In the proposed UCSFL framework, users deploy split sub-models locally, while complete models are maintained and updated at access point (AP)-side distributed processing units (DPUs), followed by a two-level aggregation procedure across DPUs and the central processing unit (CPU). Under standard machine learning (ML) assumptions, we provide a theoretical convergence analysis for UCSFL, which reveals that the AP-cluster size is a key factor influencing model training accuracy. Motivated by this result, we introduce a new performance metric, termed the latency-to-accuracy ratio, defined as the ratio of a user's per-iteration training latency to the weighted size of its AP cluster. Based on this metric, we formulate a joint optimization problem to minimize the maximum latency-to-accuracy ratio by jointly optimizing uplink power control, downlink beamforming, model splitting, and AP clustering. The resulting problem is decomposed into two sub-problems operating on different time scales, for which dedicated algorithms are developed to handle the short-term and long-term optimizations, respectively. Simulation results verify the convergence of the proposed algorithms and demonstrate that UCSFL effectively reduces the latency-to-accuracy ratio of the VGG16 model compared with baseline schemes. Moreover, the proposed framework adaptively adjusts splitting and clustering strategies in response to varying communication and computation resources. An MNIST-based handwritten digit classification example further shows that UCSFL significantly accelerates the convergence of the VGG16 model.

eess.SP↗

arXiv2025

DarkAI: Reconstructing the density, velocity and tidal field of dark matter from a DESI-like bright galaxy sample

Reconstructing the mass density, velocity, and tidal (MTV) fields of dark matter from galaxy surveys is essential for advancing our understanding of the LSS of the Universe. In this work, we present a machine learning-based framework using a UNet convolutional neural network to reconstruct the MTV fields from mock samples of the DESI bright galaxy survey within the redshift range $0.1 < z < 0.4$. Our approach accounts for realistic observational effects, including geometric selection, flux-limited data, and redshift space distortion (RSD) effects, thereby improving the fidelity of the reconstructed fields. Testing on mock galaxy catalogs generated from the Jiutian N-body simulation, our method achieves significant accuracy level. The reconstructed density field exhibits strong consistency with the true field, effectively eliminating most RSD effects and achieving a cross-correlation power spectrum coefficient greater than 0.985 on scales with $k < 0.1 \, h \, \mathrm{Mpc}^{-1}$. The velocity field reconstruction accurately captures large-scale coherent flows and small-scale turbulent features, exhibiting slopes of grid-to-grid relationships close to unity and scatter below $\sim$100 $\mathrm{km} \, \mathrm{s}^{-1}$. Additionally, the tidal field is reconstructed without bias, successfully recovering the features of the large-scale cosmic web, including clusters, filaments, sheets, and voids. Our results confirm that the proposed framework effectively captures the large-scale distribution and dynamics of dark matter while addressing key systematic challenges. These advancements provide a reliable and robust tool for analyzing current and future galaxy surveys, paving the way for new insights into cosmic structure formation and evolution.

astro-ph.CO↗

arXiv2025

DMS-Net:Dual-Modal Multi-Scale Siamese Network for Binocular Fundus Image Classification

Ophthalmic diseases pose a significant global health burden. However, traditional diagnostic methods and existing monocular image-based deep learning approaches often overlook the pathological correlations between the two eyes. In practical medical robotic diagnostic scenarios, paired retinal images (binocular fundus images) are frequently required as diagnostic evidence. To address this, we propose DMS-Net-a dual-modal multi-scale siamese network for binocular retinal image classification. The framework employs a weight-sharing siamese ResNet-152 architecture to concurrently extract deep semantic features from bilateral fundus images. To tackle challenges like indistinct lesion boundaries and diffuse pathological distributions, we introduce the OmniPool Spatial Integrator Module (OSIM), which achieves multi-resolution feature aggregation through multi-scale adaptive pooling and spatial attention mechanisms. Furthermore, the Calibrated Analogous Semantic Fusion Module (CASFM) leverages spatial-semantic recalibration and bidirectional attention mechanisms to enhance cross-modal interaction, aggregating modality-agnostic representations of fundus structures. To fully exploit the differential semantic information of lesions present in bilateral fundus features, we introduce the Cross-Modal Contrastive Alignment Module (CCAM). Additionally, to enhance the aggregation of lesion-correlated semantic information, we introduce the Cross-Modal Integrative Alignment Module (CIAM). Evaluation on the ODIR-5K dataset demonstrates that DMS-Net achieves state-of-the-art performance with an accuracy of 82.9%, recall of 84.5%, and a Cohen's kappa coefficient of 83.2%, showcasing robust capacity in detecting symmetrical pathologies and improving clinical decision-making for ocular diseases. Code and the processed dataset will be released subsequently.

cs.CV↗

arXiv2025

DiffDecompose: Layer-Wise Decomposition of Alpha-Composited Images via Diffusion Transformers

Diffusion models have recently motivated great success in many generation tasks like object removal. Nevertheless, existing image decomposition methods struggle to disentangle semi-transparent or transparent layer occlusions due to mask prior dependencies, static object assumptions, and the lack of datasets. In this paper, we delve into a novel task: Layer-Wise Decomposition of Alpha-Composited Images, aiming to recover constituent layers from single overlapped images under the condition of semi-transparent/transparent alpha layer non-linear occlusion. To address challenges in layer ambiguity, generalization, and data scarcity, we first introduce AlphaBlend, the first large-scale and high-quality dataset for transparent and semi-transparent layer decomposition, supporting six real-world subtasks (e.g., translucent flare removal, semi-transparent cell decomposition, glassware decomposition). Building on this dataset, we present DiffDecompose, a diffusion Transformer-based framework that learns the posterior over possible layer decompositions conditioned on the input image, semantic prompts, and blending type. Rather than regressing alpha mattes directly, DiffDecompose performs In-Context Decomposition, enabling the model to predict one or multiple layers without per-layer supervision, and introduces Layer Position Encoding Cloning to maintain pixel-level correspondence across layers. Extensive experiments on the proposed AlphaBlend dataset and public LOGO dataset verify the effectiveness of DiffDecompose. The code and dataset will be available upon paper acceptance. Our code will be available at: https://github.com/Wangzt1121/DiffDecompose.

cs.CV↗

arXiv2025

Extreme Solar Storm Reveals Causal Interactions in Space Weather

Solar storms perturb Earth's magnetosphere, triggering geomagnetic storms that threaten space-based systems and infrastructure. Despite advances in spaceborne and ground-based observations, the causal chain driving solar-magnetosphere-ionosphere dynamics remains elusive due to multiphysics coupling, nonlinearity, and cross-scale complexity. This study presents an information-theoretic framework to decipher interaction mechanisms in extreme solar geomagnetic storms across intensity levels within space weather causal chains, using 1980-2024 datasets. Unexpectedly, we uncover auroral spatial causality patterns associated with space weather threats in the Arctic during May 2024 extreme storms. By integrating causal consistency constraints into spatiotemporal modeling, SolarAurora outperforms existing frameworks, achieving superior accuracy in forecasting May/October 2024 events. These results advance understanding of space weather dynamics and establish a promising framework for scientific discovery and forecasting extreme space weather events.

physics.space-ph↗

arXiv2025

A Time-Triggered Communication Method Based on Urgency-Based Scheduler in Time-Sensitive Networking

The development of the automotive industry and automation has led to a growing demand for time-critical systems to have low latency and jitter for critical traffic. To address this issue, the IEEE 802.1 Time-Sensitive Networking (TSN) task group proposed the Time-Aware Shaper (TAS) to implement Time-Triggered (TT) communication, enabling deterministic transmission by assigning specific time windows to each stream. While Fixed Routing and Waiting-Allowed (FR-WA) scheduling algorithms offer flexibility, they suffer from inefficiencies in solution time and scalability. This study analyzes TAS implementation challenges, emphasizing how network scale expansion increases computational constraints. We propose an Urgency-Based Scheduler method (TT-UBS) to address these limitations to enhance deterministic transmission and computational efficiency under anomalies. A novel scheduling algorithm for TT-UBS parameter determination is developed, alongside simulations and comparative evaluations. Results show that TT-UBS guarantees deterministic traffic delivery while reducing solution time by 98.22% in test scenarios compared to traditional approaches. The methodology is extended to other scheduling algorithms to assess efficiency improvements. This advancement supports TSN's application in mission-critical systems by optimizing time-triggered communication performance and enabling reliable network deployment. The framework demonstrates significant potential for real-time in-vehicle networks requiring latency and jitter control.

cs.NI↗

arXiv2025

DeepForm: Reasoning Large Language Model for Communication System Formulation

Communication system formulation is critical for advancing 6G and future wireless technologies, yet it remains a complex, expertise-intensive task. While Large Language Models (LLMs) offer potential, existing general-purpose models often lack the specialized domain knowledge, nuanced reasoning capabilities, and access to high-quality, domain-specific training data required for adapting a general LLM into an LLM specially for communication system formulation. To bridge this gap, we introduce DeepForm, the first reasoning LLM specially for automated communication system formulation. We propose the world-first large-scale, open-source dataset meticulously curated for this domain called Communication System Formulation Reasoning Corpus (CSFRC). Our framework employs a two-stage training strategy: first, Supervised Fine-Tuning (SFT) with Chain-of-Thought (CoT) data to distill domain knowledge; second, a novel rule-based Reinforcement Learning (RL) algorithm, C-ReMax based on ReMax, to cultivate advanced modeling capabilities and elicit sophisticated reasoning patterns like self-correction and verification. Extensive experiments demonstrate that our model achieves state-of-the-art performance, significantly outperforming larger proprietary LLMs on diverse senerios. We will release related resources to foster further research in this area after the paper is accepted.

cs.LG↗

arXiv2025

Stochastic Trajectory Optimization for Robotic Skill Acquisition From a Suboptimal Demonstration

Learning from Demonstration (LfD) has emerged as a crucial method for robots to acquire new skills. However, when given suboptimal task trajectory demonstrations with shape characteristics reflecting human preferences but subpar dynamic attributes such as slow motion, robots not only need to mimic the behaviors but also optimize the dynamic performance. In this work, we leverage optimization-based methods to search for a superior-performing trajectory whose shape is similar to that of the demonstrated trajectory. Specifically, we use Dynamic Time Warping (DTW) to quantify the difference between two trajectories and combine it with additional performance metrics, such as collision cost, to construct the cost function. Moreover, we develop a multi-policy version of the Stochastic Trajectory Optimization for Motion Planning (STOMP), called MSTOMP, which is more stable and robust to parameter changes. To deal with the jitter in the demonstrated trajectory, we further utilize the gain-controlling method in the frequency domain to denoise the demonstration and propose a computationally more efficient metric, called Mean Square Error in the Spectrum (MSES), that measures the trajectories' differences in the frequency domain. We also theoretically highlight the connections between the time domain and the frequency domain methods. Finally, we verify our method in both simulation experiments and real-world experiments, showcasing its improved optimization performance and stability compared to existing methods.

cs.RO↗

arXiv2025

Cosmological distance forecasts for the CSST Galaxy Survey using BAO peaks

The measurement of cosmological distances using baryon acoustic oscillations (BAO) is crucial for studying the universe's expansion. The Chinese Space Station Telescope (CSST) galaxy redshift survey, with its vast volume and sky coverage, provides an opportunity to address key challenges in cosmology. However, redshift uncertainties in galaxy surveys can degrade both angular and radial distance estimates. In this study, we forecast the precision of BAO distance measurements using mock CSST galaxy samples, applying a two-point correlation function (2PCF) wedge approach to mitigate redshift errors. We simulate redshift uncertainties of $σ_0 = 0.003$ and $σ_0 = 0.006$, representative of expected CSST errors, and examine their effects on the BAO peak and distance scaling factors, $α_\perp$ and $α_\parallel$, across redshift bins within $0.0 < z \leqslant 1.0$. The wedge 2PCF method proves more effective in detecting the BAO peak compared to the monopole 2PCF, particularly for $σ_0 = 0.006$. Constraints on the BAO peaks show that $α_\perp$ is well constrained around 1.0, regardless of $σ_0$, with precision between 1% and 3% across redshift bins. In contrast, $α_\parallel$ measurements are more sensitive to increases in $σ_0$. For $σ_0 = 0.003$, the results remain close to the fiducial value, with uncertainties ranging between 4% and 9%; for $σ_0 = 0.006$, significant deviations from the fiducial value are observed. We also study the ability to measure parameters $(Ω_m, H_0r_\mathrm{d})$ using distance measurements, proving robust constraints as a cosmological probe under CSST-like redshift uncertainties.

astro-ph.CO↗

arXiv2024

Schedulability Analysis in Time-Sensitive Networking: A Systematic Literature Review

Time-Sensitive Networking (TSN) is a set of standards that provide low-latency, high-reliability guarantees for the transmission of traffic in networks, and it is becoming an accepted solution for complex time-critical systems such as those in industrial automation and the automotive. In time-critical systems, it is essential to verify the timing predictability of the system, and the application of scheduling mechanisms in TSN can also bring about changes in system timing. Therefore, schedulability analysis techniques can be used to verify that the system is scheduled according to the scheduling mechanism and meets the timing requirements. In this paper, we provide a clear overview of the state-of-the-art works on the topic of schedulability analysis in TSN in an attempt to clarify the purpose of schedulability analysis, categorize the methods of schedulability analysis and compare their respective strengths and weaknesses, point out the scheduling mechanisms under analyzing and the corresponding traffic classes, clarify the network scenarios constructed during the evaluation and list the challenges and directions still needing to be worked on in schedulability analysis in TSN. To this end, we conducted a systematic literature review and finally identified 123 relevant research papers published in major conferences and journals in the past 15 years. Based on a comprehensive review of the relevant literature, we have identified several key findings and emphasized the future challenges in schedulability analysis for TSN.

cs.NI↗

arXiv2024

A Centralized Discovery-Based Method for Integrating Data Distribution Service and Time-Sensitive Networking in In-Vehicle Networks

As the electronic and electrical architecture (E/EA) of intelligent and connected vehicles (ICVs) evolves, traditional distributed and signal-oriented architectures are being replaced by centralized, service-oriented architectures (SOA). This new generation of E/EA demands in-vehicle networks (IVNs) that offer high bandwidth, real-time, reliability, and service-oriented. data distribution service (DDS) and time-sensitive networking (TSN) are increasingly adopted to address these requirements. However, research on the integrated deployment of DDS and TSN in automotive applications is still in its infancy. This paper presents a DDS over TSN (DoT) communication architecture based on the centralized discovery architecture (CDA). First, a lightweight DDS implementation (FastDDS-lw) is developed for resource-constrained in-vehicle devices. Next, a DDS flow identification algorithm (DFIA) based on the CDA is introduced to identify potential DDS flows during the discovery phase automatically. Finally, the DoT communication architecture is designed, incorporating FastDDS-lw and DFIA. Experimental results show that the DoT architecture significantly reduces end-to-end latency and jitter for critical DDS flows compared to traditional Ethernet. Additionally, DoT provides an automated network configuration method that completes within a few tens of milliseconds.

cs.NI↗

arXiv2024

Imbalanced Graph-Level Anomaly Detection via Counterfactual Augmentation and Feature Learning

Graph-level anomaly detection (GLAD) has already gained significant importance and has become a popular field of study, attracting considerable attention across numerous downstream works. The core focus of this domain is to capture and highlight the anomalous information within given graph datasets. In most existing studies, anomalies are often the instances of few. The stark imbalance misleads current GLAD methods to focus on learning the patterns of normal graphs more, further impacting anomaly detection performance. Moreover, existing methods predominantly utilize the inherent features of nodes to identify anomalous graph patterns which is approved suboptimal according to our experiments. In this work, we propose an imbalanced GLAD method via counterfactual augmentation and feature learning. Specifically, we first construct anomalous samples based on counterfactual learning, aiming to expand and balance the datasets. Additionally, we construct a module based on Graph Neural Networks (GNNs), which allows us to utilize degree attributes to complement the inherent attribute features of nodes. Then, we design an adaptive weight learning module to integrate features tailored to different datasets effectively to avoid indiscriminately treating all features as equivalent. Furthermore, extensive baseline experiments conducted on public datasets substantiate the robustness and effectiveness. Besides, we apply the model to brain disease datasets, which can prove the generalization capability of our work. The source code of our work is available online.

cs.LG↗

arXiv2024

21-cm foreground removal using AI and frequency-difference technique

The deep learning technique has been employed in removing foreground contaminants from 21 cm intensity mapping, but its effectiveness is limited by the large dynamic range of the foreground amplitude. In this study, we develop a novel foreground removal technique grounded in U-Net networks. The essence of this technique lies in introducing an innovative data preprocessing step specifically, utilizing the temperature difference between neighboring frequency bands as input, which can substantially reduce the dynamic range of foreground amplitudes by approximately two orders of magnitude. This reduction proves to be highly advantageous for the U-Net foreground removal. We observe that the HI signal can be reliably recovered, as indicated by the cross-correlation power spectra showing unity agreement at the scale of $k < 0.3 h^{-1}$Mpc in the absence of instrumental effects. Moreover, accounting for the systematic beam effects, our reconstruction displays consistent auto-correlation and cross-correlation power spectrum ratios at the $1σ$ level across scales $k \lesssim 0.1 h^{-1}$Mpc, with only a 10% reduction observed in the cross-correlation power spectrum at $k\simeq0.2 h^{-1}$Mpc. The effects of redshift-space distortion are also reconstructed successfully, as evidenced by the quadrupole power spectra matching. In comparison, our method outperforms the traditional Principal Component Analysis method, which derived cross-correlation ratios are underestimated by around 60%. We simulated various white noise levels in the map and found that the mean cross-correlation ratio $\bar{R}_\mathrm{cross} \gtrsim 0.8$ when the level of the thermal noise is smaller than or equal to that of the HI signal. We conclude that the proposed frequency-difference technique can significantly enhance network performance by reducing the amplitude range of foregrounds and aiding in the prevention of HI loss.

astro-ph.CO↗

arXiv2024

Learning from Sparse Offline Datasets via Conservative Density Estimation

Offline reinforcement learning (RL) offers a promising direction for learning policies from pre-collected datasets without requiring further interactions with the environment. However, existing methods struggle to handle out-of-distribution (OOD) extrapolation errors, especially in sparse reward or scarce data settings. In this paper, we propose a novel training algorithm called Conservative Density Estimation (CDE), which addresses this challenge by explicitly imposing constraints on the state-action occupancy stationary distribution. CDE overcomes the limitations of existing approaches, such as the stationary distribution correction method, by addressing the support mismatch issue in marginal importance sampling. Our method achieves state-of-the-art performance on the D4RL benchmark. Notably, CDE consistently outperforms baselines in challenging tasks with sparse rewards or insufficient data, demonstrating the advantages of our approach in addressing the extrapolation error problem in offline RL.

cs.LG↗

arXiv2023

(DarkAI) Mapping the large-scale density field of dark matter using artificial intelligence

Herein, we present a deep-learning technique for reconstructing the dark-matter density field from the redshift-space distribution of dark-matter halos. We built a UNet-architecture neural network and trained it using the COmoving Lagrangian Acceleration fast simulation, which is an approximation of the N-body simulation with $512^3$ particles in a box size of 500 Mpc $h^{-1}$. Further, we tested the resulting UNet model not only with training-like test samples but also with standard N-body simulations, such as the Jiutian simulation with $6144^3$ particles in a box size of 1000 Mpc $h^{-1}$ and the ELUCID simulation, which has a different cosmology. The real-space dark-matter density fields in the three simulations can be reconstructed reliably with only a small reduction of the cross-correlation power spectrum at 1% and 10% levels at $k=0.1$ and $0.3~h\mathrm{Mpc^{-1}}$, respectively. The reconstruction clearly helps to correct for redshift-space distortions and is unaffected by the different cosmologies between the training (Planck2018) and test samples (WMAP5). Furthermore, we tested the application of the UNet-reconstructed density field to obtain the velocity \& tidal field and found that this approach provides better results compared to the traditional approach based on the linear bias model, showing a 12.2% improvement in the correlation slope and a 21.1% reduction in the scatter between the predicted and true velocities. Thus, our method is highly efficient and has excellent extrapolation reliability beyond the training set. This provides an ideal solution for determining the three-dimensional underlying density field from the plentiful galaxy survey data.

astro-ph.CO↗

arXiv2023

A serial dual-channel library occupancy detection system based on Faster RCNN

The phenomenon of seat occupancy in university libraries is a prevalent issue. However, existing solutions, such as software-based seat reservations and sensors-based occupancy detection, have proven to be inadequate in effectively addressing this problem. In this study, we propose a novel approach: a serial dual-channel object detection model based on Faster RCNN. This model is designed to discern all instances of occupied seats within the library and continuously update real-time information regarding seat occupancy status. To train the neural network, a distinctive dataset is utilized, which blends virtual images generated using Unreal Engine 5 (UE5) with real-world images. Notably, our test results underscore the remarkable performance uplift attained through the application of self-generated virtual datasets in training Convolutional Neural Networks (CNNs), particularly within specialized scenarios. Furthermore, this study introduces a pioneering detection model that seamlessly amalgamates the Faster R-CNN-based object detection framework with a transfer learning-based object classification algorithm. This amalgamation not only significantly curtails the computational resources and time investments needed for neural network training but also considerably heightens the efficiency of single-frame detection rates. Additionally, a user-friendly web interface and a mobile application have been meticulously developed, constituting a computer vision-driven platform for detecting seat occupancy within library premises. Noteworthy is the substantial enhancement in seat occupancy recognition accuracy, coupled with a reduction in computational resources required for neural network training, collectively contributing to a considerable amplification in the overall efficiency of library seat management.

cs.CV↗