arXiv subjects

Ao Yang

Publications and source records attributed to Ao Yang.

11 recordsLinked to original sources

arXiv2026

Relaxing Coherence Requirements on Laser Sources for Nanoscopy through Optical Fiber Technique

High numerical-aperture (NA) focusing of cylindrical vector beams (CVBs) typically requires costly, high-quality lasers to supply a stable vector pupil field. Here, we introduce fiber-conditioned vectorial nanofocusing, in which an optical-fiber-based mode-selective coupler projects a strongly distorted beam of a diode that costs two orders of magnitude less than the reference laser into radially or azimuthally polarized CVBs. A three-tolerance analysis clarifies its operating principle by establishing separate requirements for spatial state, temporal coherence and residual wavefront error. The resulting fields reproduce expected high-NA focal signatures and generate orientation-sensitive single-molecule excitation patterns consistent with reference-laser measurements, lowering the barrier to structured-light experiments.

physics.optics↗

arXiv2026

Dynamical Poles in Non-Hermitian Impurity Scattering

In Hermitian impurity scattering, each isolated late-time exponential is the fingerprint of a bound state. We show that this correspondence breaks down in non-Hermitian bands. For a single impurity in a non-Hermitian lattice, the late-time signal is controlled by isolated complex frequencies selected by the analytic continuation of the Green's function relevant to real-time dynamics, which we term dynamical poles (DPs). DPs need not coincide with static bound states: one may appear without any bound-state counterpart, while a static bound state may be dynamically invisible. The remainder of the signal is an incoherent background set by complex continuum edges. Our results establish that the real-time analytic structure of the Green's function, not the static eigenvalue problem alone, organizes non-Hermitian impurity scattering.

cond-mat.mes-hall↗

arXiv2026

R2G: A Multi-View Circuit Graph Benchmark Suite from RTL to GDSII

Graph neural networks (GNNs) are increasingly applied to physical design tasks such as congestion prediction and wirelength estimation, yet progress is hindered by inconsistent circuit representations and the absence of controlled evaluation protocols. We present R2G (RTL-to-GDSII), a multi-view circuit-graph benchmark suite that standardizes five stage-aware views with information parity (every view encodes the same attribute set, differing only in where features attach) over 30 open-source IP cores (up to $10^6$ nodes/edges). R2G provides an end-to-end DEF-to-graph pipeline spanning synthesis, placement, and routing stages, together with loaders, unified splits, domain metrics, and reproducible baselines. By decoupling representation choice from model choice, R2G isolates a confound that prior EDA and graph-ML benchmarks leave uncontrolled. In systematic studies with GINE, GAT, and ResGatedGCN, we find: (i) view choice dominates model choice, with Test R$^2$ varying by more than 0.3 across representations for a fixed GNN; (ii) node-centric views generalize best across both placement and routing; and (iii) decoder-head depth (3--4 layers) is the primary accuracy driver, turning divergent training into near-perfect predictions (R$^2$$>$0.99). Code and datasets are available at https://github.com/ShenShan123/R2G.

cs.CV↗

arXiv2025

Attention Distance: A Novel Metric for Directed Fuzzing with Large Language Models

In the domain of software security testing, Directed Grey-Box Fuzzing (DGF) has garnered widespread attention for its efficient target localization and excellent detection performance. However, existing approaches measure only the physical distance between seed execution paths and target locations, overlooking logical relationships among code segments. This omission can yield redundant or misleading guidance in complex binaries, weakening DGF's real-world effectiveness. To address this, we introduce \textbf{attention distance}, a novel metric that leverages a large language model's contextual analysis to compute attention scores between code elements and reveal their intrinsic connections. Under the same AFLGo configuration -- without altering any fuzzing components other than the distance metric -- replacing physical distances with attention distances across 38 real vulnerability reproduction experiments delivers a \textbf{3.43$\times$} average increase in testing efficiency over the traditional method. Compared to state-of-the-art directed fuzzers DAFL and WindRanger, our approach achieves \textbf{2.89$\times$} and \textbf{7.13$\times$} improvements, respectively. To further validate the generalizability of attention distance, we integrate it into DAFL and WindRanger, where it also consistently enhances their original performance. All related code and datasets are publicly available at https://github.com/TheBinKing/Attention\_Distance.git.

cs.SE↗

arXiv2025

Argus: A Multi-Agent Sensitive Information Leakage Detection Framework Based on Hierarchical Reference Relationships

Sensitive information leakage in code repositories has emerged as a critical security challenge. Traditional detection methods that rely on regular expressions, fingerprint features, and high-entropy calculations often suffer from high false-positive rates. This not only reduces detection efficiency but also significantly increases the manual screening burden on developers. Recent advances in large language models (LLMs) and multi-agent collaborative architectures have demonstrated remarkable potential for tackling complex tasks, offering a novel technological perspective for sensitive information detection. In response to these challenges, we propose Argus, a multi-agent collaborative framework for detecting sensitive information. Argus employs a three-tier detection mechanism that integrates key content, file context, and project reference relationships to effectively reduce false positives and enhance overall detection accuracy. To comprehensively evaluate Argus in real-world repository environments, we developed two new benchmarks, one to assess genuine leak detection capabilities and another to evaluate false-positive filtering performance. Experimental results show that Argus achieves up to 94.86% accuracy in leak detection, with a precision of 96.36%, recall of 94.64%, and an F1 score of 0.955. Moreover, the analysis of 97 real repositories incurred a total cost of only 2.2$. All code implementations and related datasets are publicly available at https://github.com/TheBinKing/Argus-Guard for further research and application.

cs.CR↗

arXiv2025

MCPGuard : Automatically Detecting Vulnerabilities in MCP Servers

The Model Context Protocol (MCP) has emerged as a standardized interface enabling seamless integration between Large Language Models (LLMs) and external data sources and tools. While MCP significantly reduces development complexity and enhances agent capabilities, its openness and extensibility introduce critical security vulnerabilities that threaten system trustworthiness and user data protection. This paper systematically analyzes the security landscape of MCP-based systems, identifying three principal threat categories: (1) agent hijacking attacks stemming from protocol design deficiencies; (2) traditional web vulnerabilities in MCP servers; and (3) supply chain security. To address these challenges, we comprehensively survey existing defense strategies, examining both proactive server-side scanning approaches, ranging from layered detection pipelines and agentic auditing frameworks to zero-trust registry systems, and runtime interaction monitoring solutions that provide continuous oversight and policy enforcement. Our analysis reveals that MCP security fundamentally represents a paradigm shift where the attack surface extends from traditional code execution to semantic interpretation of natural language metadata, necessitating novel defense mechanisms tailored to this unique threat model.

cs.CR↗

arXiv2025

RefleXGen:The unexamined code is not worth using

Security in code generation remains a pivotal challenge when applying large language models (LLMs). This paper introduces RefleXGen, an innovative method that significantly enhances code security by integrating Retrieval-Augmented Generation (RAG) techniques with guided self-reflection mechanisms inherent in LLMs. Unlike traditional approaches that rely on fine-tuning LLMs or developing specialized secure code datasets - processes that can be resource-intensive - RefleXGen iteratively optimizes the code generation process through self-assessment and reflection without the need for extensive resources. Within this framework, the model continuously accumulates and refines its knowledge base, thereby progressively improving the security of the generated code. Experimental results demonstrate that RefleXGen substantially enhances code security across multiple models, achieving a 13.6% improvement with GPT-3.5 Turbo, a 6.7% improvement with GPT-4o, a 4.5% improvement with CodeQwen, and a 5.8% improvement with Gemini. Our findings highlight that improving the quality of model self-reflection constitutes an effective and practical strategy for strengthening the security of AI-generated code.

cs.SE↗

arXiv2025

RA-Gen: A Controllable Code Generation Framework Using ReAct for Multi-Agent Task Execution

Code generation models based on large language models (LLMs) have gained wide adoption, but challenges remain in ensuring safety, accuracy, and controllability, especially for complex tasks. Existing methods often lack dynamic integration of external tools, transparent reasoning, and user control over safety. To address these issues, we propose a controllable code generation framework utilizing the ReAct paradigm for multi-agent task execution. This framework is a multi-agent system designed to enable efficient, precise, and interpretable code generation through dynamic interactions between LLMs and external resources. The framework adopts a collaborative architecture comprising four specialized agents: a Planner for task decomposition, a Searcher that leverages the ReAct framework for reasoning and tool integration, a CodeGen agent for accurate code generation, and an Extractor for structured data retrieval. The ReAct-based Searcher alternates between generating reasoning traces and executing actions, facilitating seamless integration of internal knowledge with external tools (such as search engines) to enhance accuracy and user control. Experimental results show the framework's effectiveness across multiple languages, achieving a 94.8% security rate on the SVEN dataset with CodeQL, outperforming existing approaches. Its transparent reasoning process fosters user trust and improves controllability.

cs.SE↗

arXiv2025

Algebraic States in Continuum in $ d\gt 1$ Dimensional Non-Hermitian Systems

We report the existence of algebraically localized eigenstates embedded within the continuum spectrum of 2D non-Hermitian systems with a single impurity. These modes, which we term algebraic states in continuum (AICs), decay algebraically as $1/|r|$ from the impurity site, and their energies lie within the bulk continuum spectrum under periodic boundary conditions. We analytically derive the threshold condition for the impurity strength required to generate such states. Remarkably, AICs are forbidden in Hermitian systems and in 1D non-Hermitian systems, making them unique to non-Hermitian systems in two and higher dimensions. To detect AICs, we introduce a local density of states as an experimental observable, which is readily accessible in photonic/acoustic platforms.

cond-mat.mes-hall↗

arXiv2025

PiCo: Jailbreaking Multimodal Large Language Models via Pictorial Code Contextualization

Multimodal Large Language Models (MLLMs), which integrate vision and other modalities into Large Language Models (LLMs), significantly enhance AI capabilities but also introduce new security vulnerabilities. By exploiting the vulnerabilities of the visual modality and the long-tail distribution characteristic of code training data, we present PiCo, a novel jailbreaking framework designed to progressively bypass multi-tiered defense mechanisms in advanced MLLMs. PiCo employs a tier-by-tier jailbreak strategy, using token-level typographic attacks to evade input filtering and embedding harmful intent within programming context instructions to bypass runtime monitoring. To comprehensively assess the impact of attacks, a new evaluation metric is further proposed to assess both the toxicity and helpfulness of model outputs post-attack. By embedding harmful intent within code-style visual instructions, PiCo achieves an average Attack Success Rate (ASR) of 84.13% on Gemini-Pro Vision and 52.66% on GPT-4, surpassing previous methods. Experimental results highlight the critical gaps in current defenses, underscoring the need for more robust strategies to secure advanced MLLMs.

cs.CR↗

arXiv2024

Tailoring Bound State Geometry in High-Dimensional Non-Hermitian Systems

It is generally believed that the non-Hermitian effect (NHSE), due to its non-reciprocal nature, creates barriers for the appearance of impurity bound states. In this paper, we find that in two and higher dimensions, the presence of geometry-dependent skin effect eliminates this barrier such that even an infinitesimal impurity potential can confine bound states in this type of non-Hermitian systems. By examining bound states around Bloch saddle points, we find that non-Hermiticity can disrupt the isotropy of bound states, resulting in concave dumbbell-shaped bound states. Our work reveals a geometry transition of bound state between concavity and convexity in high-dimensional non-Hermitian systems, offering theoretical insights for the experimental manipulation of bound states.

cond-mat.mes-hall↗