arXiv subjects

Zhiqiang Wang

Publications and source records attributed to Zhiqiang Wang.

At least 55 records · Page 3Linked to original sources

arXiv2025

VisMoDAl: Visual Analytics for Evaluating and Improving Corruption Robustness of Vision-Language Models

Vision-language (VL) models have shown transformative potential across various critical domains due to their capability to comprehend multi-modal information. However, their performance frequently degrades under distribution shifts, making it crucial to assess and improve robustness against real-world data corruption encountered in practical applications. While advancements in VL benchmark datasets and data augmentation (DA) have contributed to robustness evaluation and improvement, there remain challenges due to a lack of in-depth comprehension of model behavior as well as the need for expertise and iterative efforts to explore data patterns. Given the achievement of visualization in explaining complex models and exploring large-scale data, understanding the impact of various data corruption on VL models aligns naturally with a visual analytics approach. To address these challenges, we introduce VisMoDAl, a visual analytics framework designed to evaluate VL model robustness against various corruption types and identify underperformed samples to guide the development of effective DA strategies. Grounded in the literature review and expert discussions, VisMoDAl supports multi-level analysis, ranging from examining performance under specific corruptions to task-driven inspection of model behavior and corresponding data slice. Unlike conventional works, VisMoDAl enables users to reason about the effects of corruption on VL models, facilitating both model behavior understanding and DA strategy formulation. The utility of our system is demonstrated through case studies and quantitative evaluations focused on corruption robustness in the image captioning task.

cs.HC↗

arXiv2025

On the reducibility of the 1d quantum harmonic oscillator with a quasi-periodic bounded potential

Using the decay along the diagonal of the matrix representing the perturbation with respect to the Hermite basis, we prove a reducibility result in $L^2(\mathbb{R})$ for the one-dimensional quantum harmonic oscillator perturbed by time quasi-periodic potential, via a KAM iteration. The potential is only bounded (no decay at infinity is required) and its derivative with respect to the spatial variable $x$ is allowed to grow at most like $|x|^δ$ when $x$ goes to infinity, where the power $δ<1$ is arbitrary.

math.DS↗

arXiv2025

PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximization

Understanding the vulnerabilities of Large Vision Language Models (LVLMs) to jailbreak attacks is essential for their responsible real-world deployment. Most previous work requires access to model gradients, or is based on human knowledge (prompt engineering) to complete jailbreak, and they hardly consider the interaction of images and text, resulting in inability to jailbreak in black box scenarios or poor performance. To overcome these limitations, we propose a Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for toxicity maximization, referred to as PBI-Attack. Our method begins by extracting malicious features from a harmful corpus using an alternative LVLM and embedding these features into a benign image as prior information. Subsequently, we enhance these features through bidirectional cross-modal interaction optimization, which iteratively optimizes the bimodal perturbations in an alternating manner through greedy search, aiming to maximize the toxicity of the generated response. The toxicity level is quantified using a well-trained evaluation model. Experiments demonstrate that PBI-Attack outperforms previous state-of-the-art jailbreak methods, achieving an average attack success rate of 92.5% across three open-source LVLMs and around 67.3% on three closed-source LVLMs. Disclaimer: This paper contains potentially disturbing and offensive content.

cs.CR↗

arXiv2025

Explaining Black-box Language Models with Knowledge Probing Systems: A Post-hoc Explanation Perspective

Pre-trained Language Models (PLMs) are trained on large amounts of unlabeled data, yet they exhibit remarkable reasoning skills. However, the trustworthiness challenges posed by these black-box models have become increasingly evident in recent years. To alleviate this problem, this paper proposes a novel Knowledge-guided Probing approach called KnowProb in a post-hoc explanation way, which aims to probe whether black-box PLMs understand implicit knowledge beyond the given text, rather than focusing only on the surface level content of the text. We provide six potential explanations derived from the underlying content of the given text, including three knowledge-based understanding and three association-based reasoning. In experiments, we validate that current small-scale (or large-scale) PLMs only learn a single distribution of representation, and still face significant challenges in capturing the hidden knowledge behind a given text. Furthermore, we demonstrate that our proposed approach is effective for identifying the limitations of existing black-box models from multiple probing perspectives, which facilitates researchers to promote the study of detecting black-box models in an explainable way.

cs.CL↗

arXiv2025

On the intersection of Cantor set with the unit circle and some sequences

For $λ\in(0,1/2)$ let $K_λ$ be the self-similar set in $\mathbb{R}$ generated by the iterated function system $\{f_0(x)=λx, f_1(x)=λx+1-λ\}$. In this paper, we investigate the intersection of the unit circle $\mathbb{S} \subset \mathbb{R}^2$ with the Cartesian product $K_λ \times K_λ$. We prove that for $λ\in(0, 2 - \sqrt{3}]$, the intersection is trivial, i.e., \[ \mathbb{S} \cap (K_λ \times K_λ) = \{(0,1), (1,0)\}. \] If $λ\in [0.330384,1/2)$, then the intersection $\mathbb{S} \cap (K_λ \times K_λ)$ is non-trivial. In particular, if $λ\in [0.407493 , 1/2)$ the intersection $\mathbb{S} \cap (K_λ \times K_λ)$ is of cardinality continuum. Furthermore, the bound $2 - \sqrt{3}$ is sharp: there exists a sequence $\{λ_n\}_{n \in \mathbb{N}}$ with $λ_n \searrow 2 - \sqrt{3}$ such that $\mathbb{S} \cap (K_{λ_n} \times K_{λ_n})$ is non-trivial for all $n\in\mathbb{N}$. This result provides a negative answer to a problem posed by Yu (2023). Our methods extend beyond the unit circle and remain effective for many nonlinear curves. By employing tools from number theory, including the quadratic reciprocity law, we analyze the intersection of Cantor sets with some sequences. A dichotomy is established in terms of the Legendre symbol associated with the digit set, revealing a fundamental arithmetic constraint governing such intersections.

math.CA↗

arXiv2025

Universal approach to light driven "superconductivity" via preformed pairs

While there are many different mechanisms which have been proposed to understand the physics behind light induced ``superconductivity", what seems to be common to the class of materials in which this is observed are strong pairing correlations, which are present in the normal state. Here we argue, that the original ideas of Eliashberg are applicable to such a pseudogap phase and that with exposure to radiation the fermions are redistributed to higher energies where they are less deleterious to pairing. What results then is a photo-induced state with dramatically enhanced number of nearly condensed fermion pairs. In this phase, because the a.c. conductivity, $σ(ω) = σ_1(ω) + i σ_2(ω)$, is dominated by the bosonic contribution, it can be computed using conventional (Aslamazov Larkin) fluctuation theory. We, thereby, observe the expected fingerprint of this photoinduced ``superconducting" state which is a $1/ω$ dependence in $σ_2$ with fits to the data of the same quality as found for the so-called photo-enhanced (Drude) conductivity scenario. Here, however, we have a microscopic understanding of the characteristic low energy scale which appears in transport and which is necessarily temperature dependent. This approach also provides insight into recent observations of concomitant diamagnetic fluctuations. Our calculations suggest that the observed light-induced phase in these strongly paired superconductors has only short range phase coherence without long range superconducting order.

cond-mat.supr-con↗

arXiv2025

Universal coherent atom-molecule oscillations in the dynamics of the unitary Bose gas near a narrow Feshbach resonance

Quench experiments on a unitary Bose gas around a broad Feshbach resonance have led to the discovery of universal dynamics. This universality is manifested in the measured atomic momentum distributions where, asymptotically, a quasi-equilibrated metastable state is found in which both the momentum distribution and the time scales are determined by the particle density. In this paper we present counterpart studies but for the case of a very narrow Feshbach resonance of $^{133}$Cs atoms having a width of 8.3 mG. In dramatic contrast to the behavior reported earlier, a rapid quench of an atomic condensate to unitarity is observed to ultimately lead to coherent oscillations involving dynamically produced condensed and non-condensed molecules and atoms. The same characteristic frequency, determined by the Feshbach coupling, is observed in all types of particles. To understand these quench dynamics and how these different particle species are created, we develop a beyond Hartree-Fock-Bogoliubov dynamical framework including a new type of cross correlation between atoms and molecules. This leads to a quantitative consistency with the measured frequency. Our results, which can be applied to the general class of bosonic superfluids associated with narrow Feshbach resonances, establish a new paradigm for universal dynamics dominated by quantum many-body interactions.

cond-mat.quant-gas↗

arXiv2025

MaPPER: Multimodal Prior-guided Parameter Efficient Tuning for Referring Expression Comprehension

Referring Expression Comprehension (REC), which aims to ground a local visual region via natural language, is a task that heavily relies on multimodal alignment. Most existing methods utilize powerful pre-trained models to transfer visual/linguistic knowledge by full fine-tuning. However, full fine-tuning the entire backbone not only breaks the rich prior knowledge embedded in the pre-training, but also incurs significant computational costs. Motivated by the recent emergence of Parameter-Efficient Transfer Learning (PETL) methods, we aim to solve the REC task in an effective and efficient manner. Directly applying these PETL methods to the REC task is inappropriate, as they lack the specific-domain abilities for precise local visual perception and visual-language alignment. Therefore, we propose a novel framework of Multimodal Prior-guided Parameter Efficient Tuning, namely MaPPER. Specifically, MaPPER comprises Dynamic Prior Adapters guided by an aligned prior, and Local Convolution Adapters to extract precise local semantics for better visual perception. Moreover, the Prior-Guided Text module is proposed to further utilize the prior for facilitating the cross-modal alignment. Experimental results on three widely-used benchmarks demonstrate that MaPPER achieves the best accuracy compared to the full fine-tuning and other PETL methods with only 1.41% tunable backbone parameters. Our code is available at https://github.com/liuting20/MaPPER.

cs.CV↗

arXiv2025

SECNEURON: Reliable and Flexible Abuse Control in Local LLMs via Hybrid Neuron Encryption

Large language models (LLMs) with diverse capabilities are increasingly being deployed in local environments, presenting significant security and controllability challenges. These locally deployed LLMs operate outside the direct control of developers, rendering them more susceptible to abuse. Existing mitigation techniques mainly designed for cloud-based LLM services are frequently circumvented or ineffective in deployer-controlled environments. We propose SECNEURON, the first framework that seamlessly embeds classic access control within the intrinsic capabilities of LLMs, achieving reliable, cost-effective, flexible, and certified abuse control for local deployed LLMs. SECNEURON employs neuron-level encryption and selective decryption to dynamically control the task-specific capabilities of LLMs, limiting unauthorized task abuse without compromising others. We first design a task-specific neuron extraction mechanism to decouple logically related neurons and construct a layered policy tree for handling coupled neurons. We then introduce a flexible and efficient hybrid encryption framework for millions of neurons in LLMs. Finally, we developed a distribution-based decrypted neuron detection mechanism on ciphertext to ensure the effectiveness of partially decrypted LLMs. We proved that SECNEURON satisfies IND-CPA Security and Collusion Resistance Security under the Task Controllability Principle. Experiments on various task settings show that SECNEURON limits unauthorized task accuracy to below 25% while keeping authorized accuracy loss with 2%. Using an unauthorized Code task example, the accuracy of abuse-related malicious code generation was reduced from 59% to 15%. SECNEURON also mitigates unauthorized data leakage, reducing PII extraction rates to below 5% and membership inference to random guesses.

cs.CR↗

arXiv2025

Learning Cascade Ranking as One Network

Cascade Ranking is a prevalent architecture in large-scale top-k selection systems like recommendation and advertising platforms. Traditional training methods focus on single-stage optimization, neglecting interactions between stages. Recent advances have introduced interaction-aware training paradigms, but still struggle to 1) align training objectives with the goal of the entire cascade ranking (i.e., end-to-end recall of ground-truth items) and 2) learn effective collaboration patterns for different stages. To address these challenges, we propose LCRON, which introduces a novel surrogate loss function derived from the lower bound probability that ground truth items are selected by cascade ranking, ensuring alignment with the overall objective of the system. According to the properties of the derived bound, we further design an auxiliary loss for each stage to drive the reduction of this bound, leading to a more robust and effective top-k selection. LCRON enables end-to-end training of the entire cascade ranking system as a unified network. Experimental results demonstrate that LCRON achieves significant improvement over existing methods on public benchmarks and industrial applications, addressing key limitations in cascade ranking training and significantly enhancing system performance.

cs.IR↗

arXiv2025

DMRL: Data- and Model-aware Reward Learning for Data Extraction

Large language models (LLMs) are inherently vulnerable to unintended privacy breaches. Consequently, systematic red-teaming research is essential for developing robust defense mechanisms. However, current data extraction methods suffer from several limitations: (1) rely on dataset duplicates (addressable via deduplication), (2) depend on prompt engineering (now countered by detection and defense), and (3) rely on random-search adversarial generation. To address these challenges, we propose DMRL, a Data- and Model-aware Reward Learning approach for data extraction. This technique leverages inverse reinforcement learning to extract sensitive data from LLMs. Our method consists of two main components: (1) constructing an introspective reasoning dataset that captures leakage mindsets to guide model behavior, and (2) training reward models with Group Relative Policy Optimization (GRPO), dynamically tuning optimization based on task difficulty at both the data and model levels. Comprehensive experiments across various LLMs demonstrate that DMRL outperforms all baseline methods in data extraction performance.

cs.LG↗

arXiv2025

A 3D Framework for Improving Low-Latency Multi-Channel Live Streaming

The advent of 5G has driven the demand for high-quality, low-latency live streaming. However, challenges such as managing the increased data volume, ensuring synchronization across multiple streams, and maintaining consistent quality under varying network conditions persist, particularly in real-time video streaming. To address these issues, we propose a novel framework that leverages 3D virtual environments within game engines (e.g., Unity 3D) to optimize multi-channel live streaming. Our approach consolidates multi-camera video data into a single stream using multiple virtual 3D canvases, significantly increasing channel amounts while reducing latency and enhancing user flexibility. For demonstration of our approach, we utilize the Unity 3D engine to integrate multiple video inputs into a single-channel stream, supporting one-to-many broadcasting, one-to-one video calling, and real-time control of video channels. By mapping video data onto a world-space canvas and capturing it via an in-world camera, we minimize redundant data transmission, achieving efficient, low-latency streaming. Our results demonstrate that this method outperforms some existing multi-channel live streaming solutions in both latency reduction and user interaction responsiveness improvement. Our live video streaming system affiliated with this paper is also open-source at https://github.com/Aizierjiang/LiveStreaming.

cs.MM↗

arXiv2025

Rational points in Cantor sets and spectral eigenvalue problem for self-similar spectral measures

Given $q\in \mathbb{N}_{\ge 3}$ and a finite set $A\subset\mathbb{Q}$, let $$K(q,A)= \bigg\{\sum_{i=1}^{\infty} \frac{a_i}{q^{i}}:a_i \in A ~\forall i\in \mathbb{N} \bigg\}.$$ For $p\in\mathbb{N}_{\ge 2}$ let $D_p\subset\mathbb{R}$ be the set of all rational numbers having a finite $p$-ary expansion. We show in this paper that for $p \in \mathbb{N}_{\ge 2}$ with $\gcd(p,q)=1$, the intersection $D_p\cap K(q, A)$ is a finite set if and only if $\dim_H K(q, A)<1$, which is also equivalent to the fact that the set $K(q, A)$ has no interiors. We apply this result to study the spectral eigenvalue problem. For a Borel probability measure $μ$ on $\mathbb{R}$, a real number $t\in \mathbb{R}$ is called a spectral eigenvalue of $μ$ if both $E(Λ) =\big\{ e^{2 π\mathrm{i} λx}: λ\in Λ\big\}$ and $E(tΛ) = \big\{ e^{2 π\mathrm{i} tλx}: λ\in Λ\big\}$ are orthonormal bases in $L^2(μ)$ for some $Λ\subset \mathbb{R}$. For any self-similar spectral measure generated by a Hadamard triple, we provide a class of spectral eigenvalues which is dense in $[0,+\infty)$, and show that every eigen-subspace associated with these spectral eigenvalues is infinite.

math.CA↗

arXiv2025

GNN-Transformer Cooperative Architecture for Trustworthy Graph Contrastive Learning

Graph contrastive learning (GCL) has become a hot topic in the field of graph representation learning. In contrast to traditional supervised learning relying on a large number of labels, GCL exploits augmentation strategies to generate multiple views and positive/negative pairs, both of which greatly influence the performance. Unfortunately, commonly used random augmentations may disturb the underlying semantics of graphs. Moreover, traditional GNNs, a type of widely employed encoders in GCL, are inevitably confronted with over-smoothing and over-squashing problems. To address these issues, we propose GNN-Transformer Cooperative Architecture for Trustworthy Graph Contrastive Learning (GTCA), which inherits the advantages of both GNN and Transformer, incorporating graph topology to obtain comprehensive graph representations. Theoretical analysis verifies the trustworthiness of the proposed method. Extensive experiments on benchmark datasets demonstrate state-of-the-art empirical performance.

cs.LG↗

arXiv2025

Adaptive$^2$: Adaptive Domain Mining for Fine-grained Domain Adaptation Modeling

Advertising systems often face the multi-domain challenge, where data distributions vary significantly across scenarios. Existing domain adaptation methods primarily focus on building domain-adaptive neural networks but often rely on hand-crafted domain information, e.g., advertising placement, which may be sub-optimal. We think that fine-grained "domain" patterns exist that are difficult to hand-craft in online advertisement. Thus, we propose Adaptive$^2$, a novel framework that first learns domains adaptively using a domain mining module by self-supervision and then employs a shared&specific network to model shared and conflicting information. As a practice, we use VQ-VAE as the domain mining module and conduct extensive experiments on public benchmarks. Results show that traditional domain adaptation methods with hand-crafted domains perform no better than single-domain models under fair FLOPS conditions, highlighting the importance of domain definition. In contrast, Adaptive$^2$ outperforms existing approaches, emphasizing the effectiveness of our method and the significance of domain mining. We also deployed Adaptive$^2$ in the live streaming scenario of Kuaishou Advertising System, demonstrating its commercial value and potential for automatic domain identification. To the best of our knowledge, Adaptive$^2$ is the first approach to automatically learn both domain identification and adaptation in online advertising, opening new research directions for this area.

cs.LG↗

arXiv2025

Poster: Long PHP webshell files detection based on sliding window attention

Webshell is a type of backdoor, and web applications are widely exposed to webshell injection attacks. Therefore, it is important to study webshell detection techniques. In this study, we propose a webshell detection method. We first convert PHP source code to opcodes and then extract Opcode Double-Tuples (ODTs). Next, we combine CodeBert and FastText models for feature representation and classification. To address the challenge that deep learning methods have difficulty detecting long webshell files, we introduce a sliding window attention mechanism. This approach effectively captures malicious behavior within long files. Experimental results show that our method reaches high accuracy in webshell detection, solving the problem of traditional methods that struggle to address new webshell variants and anti-detection techniques.

cs.CR↗

arXiv2025

The coincidence of Rényi-Parry measures for $β$-transformation

We present a complete characterization of two different non-integers with the same Rényi-Parry measure. We prove that for two non-integers $β_1 ,β_2 >1$, the Rényi-Parry measures coincide if and only if $β_1$ is the root of equation $x^2-qx-p=0$, where $p,q\in\mathbb{N}$ with $p\leq q$, and $β_2 = β_1 + 1$, which confirms a conjecture of Bertrand-Mathis in \cite[Section III]{Bertrand-1998}.

math.DS↗

arXiv2024

Adaptable and Reliable Text Classification using Large Language Models

Text classification is fundamental in Natural Language Processing (NLP), and the advent of Large Language Models (LLMs) has revolutionized the field. This paper introduces an adaptable and reliable text classification paradigm, which leverages LLMs as the core component to address text classification tasks. Our system simplifies the traditional text classification workflows, reducing the need for extensive preprocessing and domain-specific expertise to deliver adaptable and reliable text classification results. We evaluated the performance of several LLMs, machine learning algorithms, and neural network-based architectures on four diverse datasets. Results demonstrate that certain LLMs surpass traditional methods in sentiment analysis, spam SMS detection, and multi-label classification. Furthermore, it is shown that the system's performance can be further enhanced through few-shot or fine-tuning strategies, making the fine-tuned model the top performer across all datasets. Source code and datasets are available in this GitHub repository: https://github.com/yeyimilk/llm-zero-shot-classifiers.

cs.CL↗