arXiv subjects

Wei Zhao

Publications and source records attributed to Wei Zhao.

At least 163 records · Page 9Linked to original sources

arXiv2025

Zero-Shot Defense Against Toxic Images via Inherent Multimodal Alignment in LVLMs

Large Vision-Language Models (LVLMs) have made significant strides in multimodal comprehension, thanks to extensive pre-training and fine-tuning on large-scale visual datasets. However, despite their robust textual safety mechanisms, they remain vulnerable to harmful visual inputs. Existing safeguards-typically relying on pre-filtering or fine-tuning-incur high costs and diminish overall utility. To address this critical vulnerability, we introduce SafeCLIP, a lightweight method that leverages LVLMs inherent multimodal alignment for zero-shot toxic image detection. By projecting CLIPs discarded CLS token into its text space and matching it with toxic descriptors, SafeCLIP detects harmful content without any architectural changes-adding minimal latency and enabling dynamic safety corrections during inference and fine-tuning.Experiments show that SafeCLIP achieves a 66.9% defense success rate with only 3.2% false positive rate and 7.2% overhead. In contrast, state-of-the-art methods achieve 52.9% success but have a 10.7% false positive rate and 210% overhead. Our work demonstrates that leveraging inherent multimodal alignment can yield efficient, low-cost LVLM safety. Code is available at anonymous.4open.science/r/safeclip-2C01.

cs.CL↗

arXiv2025

VLAS: Vision-Language-Action Model With Speech Instructions For Customized Robot Manipulation

Vision-language-action models (VLAs) have become increasingly popular in robot manipulation for their end-to-end design and remarkable performance. However, existing VLAs rely heavily on vision-language models (VLMs) that only support text-based instructions, neglecting the more natural speech modality for human-robot interaction. Traditional speech integration methods usually involves a separate speech recognition system, which complicates the model and introduces error propagation. Moreover, the transcription procedure would lose non-semantic information in the raw speech, such as voiceprint, which may be crucial for robots to successfully complete customized tasks. To overcome above challenges, we propose VLAS, a novel end-to-end VLA that integrates speech recognition directly into the robot policy model. VLAS allows the robot to understand spoken commands through inner speech-text alignment and produces corresponding actions to fulfill the task. We also present two new datasets, SQA and CSI, to support a three-stage tuning process for speech instructions, which empowers VLAS with the ability of multimodal interaction across text, image, speech, and robot actions. Taking a step further, a voice retrieval-augmented generation (RAG) paradigm is designed to enable our model to effectively handle tasks that require individual-specific knowledge. Our extensive experiments show that VLAS can effectively accomplish robot manipulation tasks with diverse speech commands, offering a seamless and customized interaction experience.

cs.RO↗

arXiv2025

A Survey on DRL based UAV Communications and Networking: DRL Fundamentals, Applications and Implementations

Unmanned aerial vehicles (UAVs) are playing an increasingly pivotal role in modern communication networks,offering flexibility and enhanced coverage for a variety of applica-tions. However, UAV networks pose significant challenges due to their dynamic and distributed nature, particularly when dealing with tasks such as power allocation, channel assignment, caching,and task offloading. Traditional optimization techniques often struggle to handle the complexity and unpredictability of these environments, leading to suboptimal performance. This survey provides a comprehensive examination of how deep reinforcement learning (DRL) can be applied to solve these mathematical optimization problems in UAV communications and networking.Rather than simply introducing DRL methods, the focus is on demonstrating how these methods can be utilized to solve complex mathematical models of the underlying problems. We begin by reviewing the fundamental concepts of DRL, including value-based, policy-based, and actor-critic approaches. Then,we illustrate how DRL algorithms are applied to specific UAV network tasks by discussing from problem formulations to DRL implementation. By framing UAV communication challenges as optimization problems, this survey emphasizes the practical value of DRL in dynamic and uncertain environments. We also explore the strengths of DRL in handling large-scale network scenarios and the ability to continuously adapt to changes in the environment. In addition, future research directions are outlined, highlighting the potential for DRL to further enhance UAV communications and expand its applicability to more complex,multi-agent settings.

cs.NI↗

arXiv2025

Compressive Sensing Empirical Wavelet Transform for Frequency-Banded Power Measurement Considering Interharmonics

Power measurement algorithms based on Fourier transform are susceptible to errors caused by interharmonics, while wavelet transform algorithms are particularly sensitive to even harmonics due to band decomposition effects. The empirical wavelet transform (EWT) has been demonstrated to improve measurement accuracy by effectively partitioning transition bands. However, for detecting interharmonic components, the limitation of the observation time window restricts spectral resolution, thereby limiting measurement accuracy. To address this challenge, this paper proposes a Compressive Sensing Empirical Wavelet Transform (CSEWT). The approach aims to enhance frequency resolution by integrating compressive sensing with the EWT, allowing precise identification of components across different frequency bands. This enables accurate determination of the power associated with the fundamental frequency, harmonics, and interharmonics. Test results indicate that the proposed CSEWT method can significantly improve the precision of individual frequency component measurements, even under dynamic and noisy conditions.

eess.SP↗

arXiv2025

Unsupervised Entity Alignment Based on Personalized Discriminative Rooted Tree

Entity Alignment (EA) is to link potential equivalent entities across different knowledge graphs (KGs). Most existing EA methods are supervised as they require the supervision of seed alignments, i.e., manually specified aligned entity pairs. Very recently, several EA studies have made some attempts to get rid of seed alignments. Despite achieving preliminary progress, they still suffer two limitations: (1) The entity embeddings produced by their GNN-like encoders lack personalization since some of the aggregation subpaths are shared between different entities. (2) They cannot fully alleviate the distribution distortion issue between candidate KGs due to the absence of the supervised signal. In this work, we propose a novel unsupervised entity alignment approach called UNEA to address the above two issues. First, we parametrically sample a tree neighborhood rooted at each entity, and accordingly develop a tree attention aggregation mechanism to extract a personalized embedding for each entity. Second, we introduce an auxiliary task of maximizing the mutual information between the input and the output of the KG encoder, to regularize the model and prevent the distribution distortion. Extensive experiments show that our UNEA achieves a new state-of-the-art for the unsupervised EA task, and can even outperform many existing supervised EA baselines.

cs.AI↗

arXiv2025

Aligning Multiple Knowledge Graphs in a Single Pass

Entity alignment (EA) is to identify equivalent entities across different knowledge graphs (KGs), which can help fuse these KGs into a more comprehensive one. Previous EA methods mainly focus on aligning a pair of KGs, and to the best of our knowledge, no existing EA method considers aligning multiple (more than two) KGs. To fill this research gap, in this work, we study a novel problem of aligning multiple KGs and propose an effective framework named MultiEA to solve the problem. First, we embed the entities of all the candidate KGs into a common feature space by a shared KG encoder. Then, we explore three alignment strategies to minimize the distances among pre-aligned entities. In particular, we propose an innovative inference enhancement technique to improve the alignment performance by incorporating high-order similarities. Finally, to verify the effectiveness of MultiEA, we construct two new real-world benchmark datasets and conduct extensive experiments on them. The results show that our MultiEA can effectively and efficiently align multiple KGs in a single pass. We release the source codes of MultiEA at: https://github.com/kepsail/MultiEA.

cs.CL↗

arXiv2025

Multi-scale Masked Autoencoder for Electrocardiogram Anomaly Detection

Electrocardiogram (ECG) analysis is a fundamental tool for diagnosing cardiovascular conditions, yet anomaly detection in ECG signals remains challenging due to their inherent complexity and variability. We propose Multi-scale Masked Autoencoder for ECG anomaly detection (MMAE-ECG), a novel end-to-end framework that effectively captures both global and local dependencies in ECG data. Unlike state-of-the-art methods that rely on heartbeat segmentation or R-peak detection, MMAE-ECG eliminates the need for such pre-processing steps, enhancing its suitability for clinical deployment. MMAE-ECG partitions ECG signals into non-overlapping segments, with each segment assigned learnable positional embeddings. A novel multi-scale masking strategy and multi-scale attention mechanism, along with distinct positional embeddings, enable a lightweight Transformer encoder to effectively capture both local and global dependencies. The masked segments are then reconstructed using a single-layer Transformer block, with an aggregation strategy employed during inference to refine the outputs. Experimental results demonstrate that our method achieves performance comparable to state-of-the-art approaches while significantly reducing computational complexity-approximately 1/78 of the floating-point operations (FLOPs) required for inference. Ablation studies further validate the effectiveness of each component, highlighting the potential of multi-scale masked autoencoders for anomaly detection.

cs.LG↗

arXiv2025

Democratic Training Against Universal Adversarial Perturbations

Despite their advances and success, real-world deep neural networks are known to be vulnerable to adversarial attacks. Universal adversarial perturbation, an input-agnostic attack, poses a serious threat for them to be deployed in security-sensitive systems. In this case, a single universal adversarial perturbation deceives the model on a range of clean inputs without requiring input-specific optimization, which makes it particularly threatening. In this work, we observe that universal adversarial perturbations usually lead to abnormal entropy spectrum in hidden layers, which suggests that the prediction is dominated by a small number of ``feature'' in such cases (rather than democratically by many features). Inspired by this, we propose an efficient yet effective defense method for mitigating UAPs called \emph{Democratic Training} by performing entropy-based model enhancement to suppress the effect of the universal adversarial perturbations in a given model. \emph{Democratic Training} is evaluated with 7 neural networks trained on 5 benchmark datasets and 5 types of state-of-the-art universal adversarial attack methods. The results show that it effectively reduces the attack success rate, improves model robustness and preserves the model accuracy on clean samples.

cs.LG↗

arXiv2025

Comparing latent inequality with ordinal data

We propose new ways to compare two latent distributions when only ordinal data are available and without imposing parametric assumptions on the underlying continuous distributions. First, we contribute identification results. We show how certain ordinal conditions provide evidence of between-group inequality, quantified by particular quantiles being higher in one latent distribution than in the other. We also show how other ordinal conditions provide evidence of higher within-group inequality in one distribution than in the other, quantified by particular interquantile ranges being wider in one latent distribution than in the other. Second, we propose an "inner" confidence set for the quantiles that are higher for the first latent distribution. We also describe frequentist and Bayesian inference on features of the ordinal distributions relevant to our identification results. Our contributions are illustrated by empirical examples with mental health and general health.

econ.EM↗

arXiv2025

Cobra: Extending Mamba to Multi-Modal Large Language Model for Efficient Inference

In recent years, the application of multimodal large language models (MLLM) in various fields has achieved remarkable success. However, as the foundation model for many downstream tasks, current MLLMs are composed of the well-known Transformer network, which has a less efficient quadratic computation complexity. To improve the efficiency of such basic models, we propose Cobra, a linear computational complexity MLLM. Specifically, Cobra integrates the efficient Mamba language model into the visual modality. Moreover, we explore and study various modal fusion schemes to create an effective multi-modal Mamba. Extensive experiments demonstrate that (1) Cobra achieves extremely competitive performance with current computationally efficient state-of-the-art methods, e.g., LLaVA-Phi, TinyLLaVA, and MobileVLM v2, and has faster speed due to Cobra's linear sequential modeling. (2) Interestingly, the results of closed-set challenging prediction benchmarks show that Cobra performs well in overcoming visual illusions and spatial relationship judgments. (3) Notably, Cobra even achieves comparable performance to LLaVA with about 43% of the number of parameters. We will make all codes of Cobra open-source and hope that the proposed method can facilitate future research on complexity problems in MLLM. Our project page is available at: https://sites.google.com/view/cobravlm.

cs.CV↗

arXiv2025

AFed: Algorithmic Fair Federated Learning

Federated Learning (FL) has gained significant attention as it facilitates collaborative machine learning among multiple clients without centralizing their data on a server. FL ensures the privacy of participating clients by locally storing their data, which creates new challenges in fairness. Traditional debiasing methods assume centralized access to sensitive information, rendering them impractical for the FL setting. Additionally, FL is more susceptible to fairness issues than centralized machine learning due to the diverse client data sources that may be associated with group information. Therefore, training a fair model in FL without access to client local data is important and challenging. This paper presents AFed, a straightforward yet effective framework for promoting group fairness in FL. The core idea is to circumvent restricted data access by learning the global data distribution. This paper proposes two approaches: AFed-G, which uses a conditional generator trained on the server side, and AFed-GAN, which improves upon AFed-G by training a conditional GAN on the client side. We augment the client data with the generated samples to help remove bias. Our theoretical analysis justifies the proposed methods, and empirical results on multiple real-world datasets demonstrate a substantial improvement in AFed over several baselines.

cs.LG↗

arXiv2024

Outage Probability Analysis of Uplink Heterogeneous Non-terrestrial Networks: A Novel Stochastic Geometry Model

In harsh environments such as mountainous terrain, dense vegetation areas, or urban landscapes, a single type of unmanned aerial vehicles (UAVs) may encounter challenges like flight restrictions, difficulty in task execution, or increased risk. Therefore, employing multiple types of UAVs, along with satellite assistance, to collaborate becomes essential in such scenarios. In this context, we present a stochastic geometry based approach for modeling the heterogeneous non-terrestrial networks (NTNs) by using the classical binomial point process and introducing a novel point process, called Mat{é}rn hard-core cluster process (MHCCP). Our MHCCP possesses both the exclusivity and the clustering properties, thus it can better model the aircraft group composed of multiple clusters. Then, we derive closed-form expressions of the outage probability (OP) for the uplink (aerial-to-satellite) of heterogeneous NTNs. Unlike existing studies, our analysis relies on a more advanced system configuration, where the integration of beamforming and frequency division multiple access, and the shadowed-Rician (SR) fading model for interference power, are considered. The accuracy of our theoretical derivation is confirmed by Monte Carlo simulations. Our research offers fundamental insights into the system-level performance optimization of NTNs.

cs.NI↗

arXiv2024

Do Influence Functions Work on Large Language Models?

Influence functions are important for quantifying the impact of individual training data points on a model's predictions. Although extensive research has been conducted on influence functions in traditional machine learning models, their application to large language models (LLMs) has been limited. In this work, we conduct a systematic study to address a key question: do influence functions work on LLMs? Specifically, we evaluate influence functions across multiple tasks and find that they consistently perform poorly in most settings. Our further investigation reveals that their poor performance can be attributed to: (1) inevitable approximation errors when estimating the iHVP component due to the scale of LLMs, (2) uncertain convergence during fine-tuning, and, more fundamentally, (3) the definition itself, as changes in model parameters do not necessarily correlate with changes in LLM behavior. Thus, our study suggests the need for alternative approaches for identifying influential samples.

cs.CL↗

arXiv2024

Unleashing the Unseen: Harnessing Benign Datasets for Jailbreaking Large Language Models

Despite significant ongoing efforts in safety alignment, large language models (LLMs) such as GPT-4 and LLaMA 3 remain vulnerable to jailbreak attacks that can induce harmful behaviors, including through the use of adversarial suffixes. Building on prior research, we hypothesize that these adversarial suffixes are not mere bugs but may represent features that can dominate the LLM's behavior. To evaluate this hypothesis, we conduct several experiments. First, we demonstrate that benign features can be effectively made to function as adversarial suffixes, i.e., we develop a feature extraction method to extract sample-agnostic features from benign dataset in the form of suffixes and show that these suffixes may effectively compromise safety alignment. Second, we show that adversarial suffixes generated from jailbreak attacks may contain meaningful features, i.e., appending the same suffix to different prompts results in responses exhibiting specific characteristics. Third, we show that such benign-yet-safety-compromising features can be easily introduced through fine-tuning using only benign datasets. As a result, we are able to completely eliminate GPT's safety alignment in a blackbox setting through finetuning with only benign data. Our code and data is available at \url{https://github.com/suffix-maybe-feature/adver-suffix-maybe-features}.

cs.CR↗

arXiv2024

Updates on the Tsinghua Tabletop Kibble Balance

With the adoption of the revised International System of Units (SI), the Kibble balance has become a pivotal instrument for mass calibrations against the Planck constant, $h$. One of the major focuses in the Kibble balance community is prioritizing experiments that achieve both high accuracy and compactness. The Tsinghua tabletop Kibble balance experiment seeks to develop a compact, high-precision, user-friendly, cost-effective, and open-hardware apparatus for mass realization, specifically within the kilogram range. This paper reports on the progress of the Tsinghua tabletop Kibble balance project over the past two years. Various aspects of the Tsinghua tabletop system, including electrical, magnetic, mechanical, and optical components, are summarized. Key achievements, such as the construction and characterization of the magnet system, determination of absolute gravitational acceleration, investigation of a capacitor-sensor-based weighing unit, and development of a high-precision current source, are presented to provide a comprehensive understanding of the experiment's status.

physics.ins-det↗

arXiv2024

AGMixup: Adaptive Graph Mixup for Semi-supervised Node Classification

Mixup is a data augmentation technique that enhances model generalization by interpolating between data points using a mixing ratio $λ$ in the image domain. Recently, the concept of mixup has been adapted to the graph domain through node-centric interpolations. However, these approaches often fail to address the complexity of interconnected relationships, potentially damaging the graph's natural topology and undermining node interactions. Furthermore, current graph mixup methods employ a one-size-fits-all strategy with a randomly sampled $λ$ for all mixup pairs, ignoring the diverse needs of different pairs. This paper proposes an Adaptive Graph Mixup (AGMixup) framework for semi-supervised node classification. AGMixup introduces a subgraph-centric approach, which treats each subgraph similarly to how images are handled in Euclidean domains, thus facilitating a more natural integration of mixup into graph-based learning. We also propose an adaptive mechanism to tune the mixing ratio $λ$ for diverse mixup pairs, guided by the contextual similarity and uncertainty of the involved subgraphs. Extensive experiments across seven datasets on semi-supervised node classification benchmarks demonstrate AGMixup's superiority over state-of-the-art graph mixup methods. Source codes are available at \url{https://github.com/WeigangLu/AGMixup}.

cs.LG↗

arXiv2024

Quad-cascade picture of electrokinetic turbulence

Turbulence, ubiquitous in nature and across various systems, exhibits chaotic and intermittent fluctuations in space and time, defying precise prediction. For nearly a century, extensive efforts have been made to uncover the underlying universality and invariant laws from the immense disorder and chaotic nature of turbulence. While the celebrated Kolmogorov -5/3 law stands as a robust cornerstone, it falls short in capturing the diverse scaling behavior exhibited in turbulence influenced by external volume forces, like thermal convection and electrokinetic flows. This study proposes a general framework that couples the fluxes of kinetic energy and scalar variance, culminating in the formulation of a universal conservation law. This framework offers a comprehensive quad-cascade depiction of turbulence, enabling predictions that beyond the limitations of existing models. We illustrate this framework with microfluidic experiments on electrokinetic turbulence, wherein power spectra of concentration and velocity fluctuations exhibit the predicted scaling behaviors, providing remarkable agreement with theory. These findings not only deepen our understanding of the complete cascade process in turbulence driven by external volume forces but also hold promise for insights into other turbulent systems.

physics.flu-dyn↗

arXiv2024

Quantum versatility in PageRank

Quantum mechanics empowers the emergence of quantum advantages in various fields, including quantum algorithms. Quantum PageRank is a promising tool for a future quantum internet. Recently, arbitrary phase rotations (APR) have been introduced in the underlying Szegedy's quantum walk of quantum PageRank algorithm. In this work, we thoroughly study the role APR plays in quantum PageRank. We discover the versatility resulting from quantumness. Specifically, we discover the emergence of a cluster phenomenon in rankings considering the rotation phases, i.e. the existence of similar clusters in the distribution of the rankings and their fidelity with the corresponding classical PageRanks, the ranking distribution variance, the coherence and entanglement of PageRank states, and the power law parameter in the ranking distributions on a scale-free network concerning the two rotation phases. Furthermore, we propose an alternate quantum PageRank with APR which provides an extra tunnel for the analysis of PageRank. We also study the PageRank on the trackback graph of a scale-free graph for the investigation of network information traffic tracking. We demonstrate the rich cluster diversity formed in our alternate quantum PageRank, which offers a novel perspective on the quantum versatility of PageRank. Our results present the quantum-enabled perspective for PageRanking and shed light on the design and application of practical quantum PageRank algorithms.

quant-ph↗