arXiv subjects

Wei Zhao

Publications and source records attributed to Wei Zhao.

At least 145 records · Page 8Linked to original sources

arXiv2025

Analysis of Jet Dynamics and Collimation Characteristics of 0241+622 on Parsec Scales

We conducted a detailed analysis of the jet structure and dynamics of the source 0241+622 on milliarcsecond (mas) scales. We stacked images from multiple epochs to better recover the crosssection of the jet. By analyzing the relationship between jet width and distance, we observed that the jet exhibits a parabolic shape from the core, spanning a region from 0.12 to 6.1 mas. This structure suggests the acceleration and collimation processes of the jet. Beyond 18 mas from the core, the jet adopts a conical shape, and the expansion speed of the jet becomes faster within the range from 4500 to 6500 mas. We obtained the core shift of this source using five pairs of data from VLBA at 1.6 GHz to 43 GHz. Based on previous studies, through proper motion analysis of the jet components, we estimated the angle between the jet and the line of sight to be approximately 65.7°, so 1 mas corresponds to 0.95 pc (de-projected distance). We then obtained the velocity field of the source within 3.14 mas from the central black hole and found that the jet exhibits accelerated motion within this range. At approximately 6.1 mas from the core, we observed that the jet width begins to decrease, which we identified as possibly corresponding to the Bondi radius of this source. The reduction in jet width may be related to changes in the external environmental pressure, particularly within the Bondi radius, indicating that the jet dynamics and collimation characteristics are strongly influenced by the surrounding medium conditions.

astro-ph.HE↗

arXiv2025

Multiple Object Tracking in Video SAR: A Benchmark and Tracking Baseline

In the context of multi-object tracking using video synthetic aperture radar (Video SAR), Doppler shifts induced by target motion result in artifacts that are easily mistaken for shadows caused by static occlusions. Moreover, appearance changes of the target caused by Doppler mismatch may lead to association failures and disrupt trajectory continuity. A major limitation in this field is the lack of public benchmark datasets for standardized algorithm evaluation. To address the above challenges, we collected and annotated 45 video SAR sequences containing moving targets, and named the Video SAR MOT Benchmark (VSMB). Specifically, to mitigate the effects of trailing and defocusing in moving targets, we introduce a line feature enhancement mechanism that emphasizes the positive role of motion shadows and reduces false alarms induced by static occlusions. In addition, to mitigate the adverse effects of target appearance variations, we propose a motion-aware clue discarding mechanism that substantially improves tracking robustness in Video SAR. The proposed model achieves state-of-the-art performance on the VSMB, and the dataset and model are released at https://github.com/softwarePupil/VSMB.

cs.CV↗

arXiv2025

J-DDL: Surface Damage Detection and Localization System for Fighter Aircraft

Ensuring the safety and extended operational life of fighter aircraft necessitates frequent and exhaustive inspections. While surface defect detection is feasible for human inspectors, manual methods face critical limitations in scalability, efficiency, and consistency due to the vast surface area, structural complexity, and operational demands of aircraft maintenance. We propose a smart surface damage detection and localization system for fighter aircraft, termed J-DDL. J-DDL integrates 2D images and 3D point clouds of the entire aircraft surface, captured using a combined system of laser scanners and cameras, to achieve precise damage detection and localization. Central to our system is a novel damage detection network built on the YOLO architecture, specifically optimized for identifying surface defects in 2D aircraft images. Key innovations include lightweight Fasternet blocks for efficient feature extraction, an optimized neck architecture incorporating Efficient Multiscale Attention (EMA) modules for superior feature aggregation, and the introduction of a novel loss function, Inner-CIOU, to enhance detection accuracy. After detecting damage in 2D images, the system maps the identified anomalies onto corresponding 3D point clouds, enabling accurate 3D localization of defects across the aircraft surface. Our J-DDL not only streamlines the inspection process but also ensures more comprehensive and detailed coverage of large and complex aircraft exteriors. To facilitate further advancements in this domain, we have developed the first publicly available dataset specifically focused on aircraft damage. Experimental evaluations validate the effectiveness of our framework, underscoring its potential to significantly advance automated aircraft inspection technologies.

cs.CV↗

arXiv2025

Self-attention U-Net decoder for toric codes

In the NISQ era, one of the most important bottlenecks for the realization of universal quantum computation is error correction. Stabiliser code is the most recognizable type of quantum error correction code. A scalable efficient decoder is most desired for the application of the quantum error correction codes. In this work, we propose a self-attention U-Net quantum decoder (SU-NetQD) for toric code, which outperforms the minimum weight perfect matching decoder, especially in the circuit level noise environments. Specifically, with our SU-NetQD, we achieve lower logical error rates compared with MWPM and discover an increased trend of code threshold as the increase of noise bias. We obtain a high threshold of 0.231 for the extremely biased noise environment. The combination of low-level decoder and high-level decoder is the key innovation for the high accuracy of our decoder. With transfer learning mechanics, our decoder is scalable for cases with different code distances. Our decoder provides a practical tool for quantum noise analysis and promotes the practicality of quantum error correction codes and quantum computing.

quant-ph↗

arXiv2025

Selftok: Discrete Visual Tokens of Autoregression, by Diffusion, and for Reasoning

We completely discard the conventional spatial prior in image representation and introduce a novel discrete visual tokenizer: Self-consistency Tokenizer (Selftok). At its design core, we compose an autoregressive (AR) prior -- mirroring the causal structure of language -- into visual tokens by using the reverse diffusion process of image generation. The AR property makes Selftok fundamentally distinct from traditional spatial tokens in the following two key ways: - Selftok offers an elegant and minimalist approach to unify diffusion and AR for vision-language models (VLMs): By representing images with Selftok tokens, we can train a VLM using a purely discrete autoregressive architecture -- like that in LLMs -- without requiring additional modules or training objectives. - We theoretically show that the AR prior satisfies the Bellman equation, whereas the spatial prior does not. Therefore, Selftok supports reinforcement learning (RL) for visual generation with effectiveness comparable to that achieved in LLMs. Besides the AR property, Selftok is also a SoTA tokenizer that achieves a favorable trade-off between high-quality reconstruction and compression rate. We use Selftok to build a pure AR VLM for both visual comprehension and generation tasks. Impressively, without using any text-image training pairs, a simple policy gradient RL working in the visual tokens can significantly boost the visual generation benchmark, surpassing all the existing models by a large margin. Therefore, we believe that Selftok effectively addresses the long-standing challenge that visual tokens cannot support effective RL. When combined with the well-established strengths of RL in LLMs, this brings us one step closer to realizing a truly multimodal LLM. Project Page: https://selftok-team.github.io/report/.

cs.CV↗

arXiv2025

Conf-GNNRec: Quantifying and Calibrating the Prediction Confidence for GNN-based Recommendation Methods

Recommender systems based on graph neural networks perform well in tasks such as rating and ranking. However, in real-world recommendation scenarios, noise such as user misuse and malicious advertisement gradually accumulates through the message propagation mechanism. Even if existing studies mitigate their effects by reducing the noise propagation weights, the severe sparsity of the recommender system still leads to the low-weighted noisy neighbors being mistaken as meaningful information, and the prediction result obtained based on the polluted nodes is not entirely trustworthy. Therefore, it is crucial to measure the confidence of the prediction results in this highly noisy framework. Furthermore, our evaluation of the existing representative GNN-based recommendation shows that it suffers from overconfidence. Based on the above considerations, we propose a new method to quantify and calibrate the prediction confidence of GNN-based recommendations (Conf-GNNRec). Specifically, we propose a rating calibration method that dynamically adjusts excessive ratings to mitigate overconfidence based on user personalization. We also design a confidence loss function to reduce the overconfidence of negative samples and effectively improve recommendation performance. Experiments on public datasets demonstrate the validity of Conf-GNNRec in prediction confidence and recommendation performance.

cs.IR↗

arXiv2025

Unveiling the Potential of Vision-Language-Action Models with Open-Ended Multimodal Instructions

Vision-Language-Action (VLA) models have recently become highly prominent in the field of robotics. Leveraging vision-language foundation models trained on large-scale internet data, the VLA model can generate robotic actions directly from visual observations and human instructions through a single end-to-end neural network. Despite their effectiveness, current VLA models usually accept only one form of human prompting, language instructions, which may constrain their applicability in open-ended human-robot interactions. For example, a user might expect the robot to retrieve an object shown in an image, follow an instruction written on the whiteboard, or imitate a behavior demonstrated in a video, rather than relying solely on language-based descriptions. To address this gap, we introduce OE-VLA, which explores the potential of VLA models for open-ended multimodal instructions. Extensive results demonstrate that our OE-VLA not only achieves comparable performance to traditional VLA models with linguistic input but also delivers impressive results across four additional categories of open-ended tasks. The proposed methodology could significantly expand the applications of VLA models across various everyday scenarios and facilitate human-robot interaction.

cs.RO↗

arXiv2025

Quantum data generation in a denoising model with multiscale entanglement renormalization network

Quantum technology has entered the era of noisy intermediate-scale quantum (NISQ) information processing. The technological revolution of machine learning represented by generative models heralds a great prospect of artificial intelligence, and the huge amount of data processes poses a big challenge to existing computers. The generation of large quantities of quantum data will be a challenge for quantum artificial intelligence. In this work, we present an efficient noise-resistant quantum data generation method that can be applied to various types of NISQ quantum processors, where the target quantum data belongs to a certain class and our proposal enables the generation of various quantum data belonging to the target class. Specifically, we propose a quantum denoising probability model (QDM) based on a multiscale entanglement renormalization network (MERA) for the generation of quantum data. To show the feasibility and practicality of our scheme, we demonstrate the generations of the classes of GHZ-like states and W-like states with a success rate above 99%. Our MREA QDM can also be used to denoise multiple types of quantum data simultaneously. We show the success rate of denoising both GHZ-like and W-like states with a single qubit noise environment of noise level within 1/4 can approximate to be 100%, and with two other types of noise environment with noise level within 1/4 can be above 90%. Our quantum data generation scheme provides new ideas and prospects for quantum generative models in the NISQ era.

quant-ph↗

arXiv2025

Rethinking Latent Redundancy in Behavior Cloning: An Information Bottleneck Approach for Robot Manipulation

Behavior Cloning (BC) is a widely adopted visual imitation learning method in robot manipulation. Current BC approaches often enhance generalization by leveraging large datasets and incorporating additional visual and textual modalities to capture more diverse information. However, these methods overlook whether the learned representations contain redundant information and lack a solid theoretical foundation to guide the learning process. To address these limitations, we adopt an information-theoretic perspective and introduce mutual information to quantify and mitigate redundancy in latent representations. Building on this, we incorporate the Information Bottleneck (IB) principle into BC, which extends the idea of reducing redundancy by providing a structured framework for compressing irrelevant information while preserving task-relevant features. This work presents the first comprehensive study on redundancy in latent representations across various methods, backbones, and experimental settings, while extending the generalizability of the IB to BC. Extensive experiments and analyses on the CortexBench and LIBERO benchmarks demonstrate significant performance improvements with IB, underscoring the importance of reducing input data redundancy and highlighting its practical value for more practical applications. Project Page: https://baishuanghao.github.io/BC-IB.github.io.

cs.RO↗

arXiv2025

Reasoning Physical Video Generation with Diffusion Timestep Tokens via Reinforcement Learning

Despite recent progress in video generation, producing videos that adhere to physical laws remains a significant challenge. Traditional diffusion-based methods struggle to extrapolate to unseen physical conditions (eg, velocity) due to their reliance on data-driven approximations. To address this, we propose to integrate symbolic reasoning and reinforcement learning to enforce physical consistency in video generation. We first introduce the Diffusion Timestep Tokenizer (DDT), which learns discrete, recursive visual tokens by recovering visual attributes lost during the diffusion process. The recursive visual tokens enable symbolic reasoning by a large language model. Based on it, we propose the Phys-AR framework, which consists of two stages: The first stage uses supervised fine-tuning to transfer symbolic knowledge, while the second stage applies reinforcement learning to optimize the model's reasoning abilities through reward functions based on physical conditions. Our approach allows the model to dynamically adjust and improve the physical properties of generated videos, ensuring adherence to physical laws. Experimental results demonstrate that PhysAR can generate videos that are physically consistent.

cs.CV↗

arXiv2025

Partition laser assembling technique

The advancement of micro/nanofabrication techniques with high throughput, efficiency, and flexibility is critical for fields like integrated photonics, biosensing, and medical diagnostics. This study presents Partition Laser Assembling (PLA), a novel laser technique for fabricating complex micro/nanostructures akin to puzzle pieces. By dividing the target patterns described by scalable vector graphics into partitions, any structures in each partition can be fabricated via structured lights as "light stamp" through spatial light modulation. Unlike traditional direct laser writing, PLA eliminates reliance on mechanical components, avoiding step-like artifacts and ensuring smoother fabrication of complex micro/nanostructures. By seamlessly assembling basic shapes, PLA achieves intricate structures like micro artworks and metalenses with unmatched precision and resolution. Leveraging two-photon fabrication, PLA guarantees high resolution and structural integrity, positioning it as a transformative tool for nanoscale 3D printing. With applications spanning research and industry, PLA paves the way for advanced optical devices, micro/nanofabrications, and next-gen manufacturing technologies.

physics.optics↗

arXiv2025

Generative Multimodal Pretraining with Discrete Diffusion Timestep Tokens

Recent endeavors in Multimodal Large Language Models (MLLMs) aim to unify visual comprehension and generation by combining LLM and diffusion models, the state-of-the-art in each task, respectively. Existing approaches rely on spatial visual tokens, where image patches are encoded and arranged according to a spatial order (e.g., raster scan). However, we show that spatial tokens lack the recursive structure inherent to languages, hence form an impossible language for LLM to master. In this paper, we build a proper visual language by leveraging diffusion timesteps to learn discrete, recursive visual tokens. Our proposed tokens recursively compensate for the progressive attribute loss in noisy images as timesteps increase, enabling the diffusion model to reconstruct the original image at any timestep. This approach allows us to effectively integrate the strengths of LLMs in autoregressive reasoning and diffusion models in precise image generation, achieving seamless multimodal comprehension and generation within a unified framework. Extensive experiments show that we achieve superior performance for multimodal comprehension and generation simultaneously compared with other MLLMs. Project Page: https://DDT-LLaMA.github.io/.

cs.CV↗

arXiv2025

DA2Diff: Exploring Degradation-aware Adaptive Diffusion Priors for All-in-One Weather Restoration

Image restoration under adverse weather conditions is a critical task for many vision-based applications. Recent all-in-one frameworks that handle multiple weather degradations within a unified model have shown potential. However, the diversity of degradation patterns across different weather conditions, as well as the complex and varied nature of real-world degradations, pose significant challenges for multiple weather removal. To address these challenges, we propose an innovative diffusion paradigm with degradation-aware adaptive priors for all-in-one weather restoration, termed DA2Diff. It is a new exploration that applies CLIP to perceive degradation-aware properties for better multi-weather restoration. Specifically, we deploy a set of learnable prompts to capture degradation-aware representations by the prompt-image similarity constraints in the CLIP space. By aligning the snowy/hazy/rainy images with snow/haze/rain prompts, each prompt contributes to different weather degradation characteristics. The learned prompts are then integrated into the diffusion model via the designed weather specific prompt guidance module, making it possible to restore multiple weather types. To further improve the adaptiveness to complex weather degradations, we propose a dynamic expert selection modulator that employs a dynamic weather-aware router to flexibly dispatch varying numbers of restoration experts for each weather-distorted image, allowing the diffusion model to restore diverse degradations adaptively. Experimental results substantiate the favorable performance of DA2Diff over state-of-the-arts in quantitative and qualitative evaluation. Source code will be available after acceptance.

cs.CV↗

arXiv2025

Betweenness Centrality Based Dynamic Source Routing for Flying Ad Hoc Networks in Marching Formation

Designing high-performance routing protocols for flying ad hoc networks (FANETs) is challenging due to the diversity of applications and the dynamics of network topology. The existing general-purpose routing protocols for ad hoc networks often oversimplify mobility patterns and disregard the unequal importance of nodes, resulting in suboptimal routing decisions that are unsuitable for task-oriented FANETs. To break the bottleneck, in this paper we propose a betweenness centrality based dynamic source routing (BC-DSR) protocol for a flying ad hoc network (FANET) in marching formation. Firstly, we introduce a Gauss-Markov group (GMG) mobility model based on the leader-follower pattern, which accurately captures the temporal and spatial correlations of node movements in the realistic marching formation. Besides, we exploit the concept of BC defined in graph theory to measure the structural unequal importance of relay nodes, i.e., to determine link weights, in the particular marching formation topology. The path of least cost is calculated relying on a weighted directed graph constructed. The ns-3 based simulation results demonstrate that our BCDSR protocol achieves higher packet-delivery ratio and lower average end-to-end latency and routing overhead ratio than representative benchmark protocols used in FANETs, while maintaining a reasonably small network jitter.

cs.NI↗

arXiv2025

Seesaw: High-throughput LLM Inference via Model Re-sharding

To improve the efficiency of distributed large language model (LLM) inference, various parallelization strategies, such as tensor and pipeline parallelism, have been proposed. However, the distinct computational characteristics inherent in the two stages of LLM inference-prefilling and decoding-render a single static parallelization strategy insufficient for the effective optimization of both stages. In this work, we present Seesaw, an LLM inference engine optimized for throughput-oriented tasks. The key idea behind Seesaw is dynamic model re-sharding, a technique that facilitates the dynamic reconfiguration of parallelization strategies across stages, thereby maximizing throughput at both phases. To mitigate re-sharding overhead and optimize computational efficiency, we employ tiered KV cache buffering and transition-minimizing scheduling. These approaches work synergistically to reduce the overhead caused by frequent stage transitions while ensuring maximum batching efficiency. Our evaluation demonstrates that Seesaw achieves a throughput increase of up to 1.78x (1.36x on average) compared to vLLM, the most widely used state-of-the-art LLM inference engine.

cs.DC↗

arXiv2025

Metering Error Estimation of Fast-Charging Stations Using Charging Data Analytics

Accurate electric energy metering (EEM) of fast charging stations (FCSs), serving as critical infrastructure in the electric vehicle (EV) industry and as significant carriers of vehicle-to-grid (V2G) technology, is the cornerstone for ensuring fair electric energy transactions. Traditional on-site verification methods, constrained by their high costs and low efficiency, struggle to keep pace with the rapid global expansion of FCSs. In response, this paper adopts a data-driven approach and proposes the measuring performance comparison (MPC) method. By utilizing the estimation value of state-of-charge (SOC) as a medium, MPC establishes comparison chains of EEM performance of multiple FCSs. Therefore, the estimation of EEM errors for FCSs with high efficiency is enabled. Moreover, this paper summarizes the interfering factors of estimation results and establishes corresponding error models and uncertainty models. Also, a method for discriminating whether there are EEM performance defects in FCSs is proposed. Finally, the feasibility of MPC method is validated, with results indicating that for FCSs with an accuracy grade of 2\%, the discriminative accuracy exceeds 95\%. The MPC provides a viable approach for the online monitoring of EEM performance for FCSs, laying a foundation for a fair and just electricity trading market.

eess.SP↗

arXiv2025

Tally: Non-Intrusive Performance Isolation for Concurrent Deep Learning Workloads

GPU underutilization is a significant concern in many production deep learning clusters, leading to prolonged job queues and increased operational expenses. A promising solution to this inefficiency is GPU sharing, which improves resource utilization by allowing multiple workloads to execute concurrently on a single GPU. However, deploying GPU sharing in production settings faces critical obstacles due to the limitations of existing mechanisms, including high integration costs, inadequate performance isolation, and limited application compatibility. To address these issues, we introduce \emph{Tally}, a non-intrusive GPU sharing mechanism that provides robust performance isolation and comprehensive workload compatibility. The key to Tally's robust performance isolation capability lies in its fine-grained thread-block-level GPU kernel scheduling strategy, which allows the system to effectively mitigate interference caused by workload co-execution. We evaluate Tally on a diverse range of workloads and show that it incurs an average overhead of only $7.2\%$ on the $99^{th}$-percentile latency of high-priority inference tasks when executed concurrently with best-effort training workloads, compared to $188.9\%$ overhead exhibited by the state-of-the-art GPU sharing systems like TGS, while achieving over $80\%$ of TGS's system throughput.

cs.DC↗

arXiv2025

How Good Are LLMs for Literary Translation, Really? Literary Translation Evaluation with Humans and LLMs

Recent research has focused on literary machine translation (MT) as a new challenge in MT. However, the evaluation of literary MT remains an open problem. We contribute to this ongoing discussion by introducing LITEVAL-CORPUS, a paragraph-level parallel corpus containing verified human translations and outputs from 9 MT systems, which totals over 2k translations and 13k evaluated sentences across four language pairs, costing 4.5k C. This corpus enables us to (i) examine the consistency and adequacy of human evaluation schemes with various degrees of complexity, (ii) compare evaluations by students and professionals, assess the effectiveness of (iii) LLM-based metrics and (iv) LLMs themselves. Our findings indicate that the adequacy of human evaluation is controlled by two factors: the complexity of the evaluation scheme (more complex is less adequate) and the expertise of evaluators (higher expertise yields more adequate evaluations). For instance, MQM (Multidimensional Quality Metrics), a complex scheme and the de facto standard for non-literary human MT evaluation, is largely inadequate for literary translation evaluation: with student evaluators, nearly 60% of human translations are misjudged as indistinguishable or inferior to machine translations. In contrast, BWS (BEST-WORST SCALING), a much simpler scheme, identifies human translations at a rate of 80-100%. Automatic metrics fare dramatically worse, with rates of at most 20%. Our overall evaluation indicates that published human translations consistently outperform LLM translations, where even the most recent LLMs tend to produce considerably more literal and less diverse translations compared to humans.

cs.CL↗