arXiv subjects

Ting Yu

Publications and source records attributed to Ting Yu.

At least 37 records · Page 2Linked to original sources

arXiv2026

Image Prompt Reconstruction Attacks on Distributed MLLM Inference Frameworks

Distributed large language model (LLM) inference frameworks connect isolated consumer-grade devices for large-scale model inference, substantially reducing hardware constraints. However, recent studies show that intermediate embeddings transmitted among participants can leak private prompts. As LLMs evolve into multimodal LLMs (MLLMs), this risk extends beyond text: image prompts contain rich visual and semantic information, making their intermediate embeddings highly privacy-sensitive. Yet, image-prompt leakage in distributed MLLM inference remains largely unexplored. In this paper, we investigate privacy risks to input images caused by intermediate embeddings in distributed MLLM frameworks. We first analyze the information flow from image pixels to intermediate representations. Since image and text embeddings are often intertwined across MLLM layers, we design an image embedding extraction algorithm as a prerequisite for reconstruction attacks, achieving 100% extraction accuracy across almost all MLLM layers in our experiments. Building on this, we develop two passive black-box image reconstruction attacks, MPAA and IEDA, reflecting realistic threats from normal participants with limited knowledge and capability. MPAA performs fine-grained pixel-level reconstruction via patch-wise information extraction and assembly, while IEDA performs coarse-grained semantic reconstruction through embedding-guided diffusion generation. We evaluate our attacks on four representative MLLM families: Gemma 3, Phi 4 Multimodal, Qwen 2.5 VL, and Llama 4 Scout. Results show consistently superior reconstruction performance in various settings. We further analyze the effects of MoE architecture, image preprocessing, model size, and text-image dependency on attack performance. To our knowledge, this is the first study of image reconstruction attacks on MLLMs.

cs.CR↗

arXiv2026

Quantum Fisher Information and the Curvature of Entanglement

We explore the relationship between quantum Fisher information (QFI) and the negative of the second derivative of concurrence with respect to the coupling between two qubits, referred to as the curvature of entanglement (CoE). The two-qubit system serves as a minimal model to study the connection between QFI and dynamically generated entanglement in scenarios where the measured quantity is a two- or many-body coupling strength. We analyze in detail the pure-state lossless case for which general results can be inferred and we also consider a simple interaction Hamiltonian in the case of one form of loss applied to the qubits. For a two-qubit quantum probe used to estimate the coupling constant appearing in the interaction Hamiltonian we show, for certain initial conditions, that there are times such that CoE = QFI. These times can be associated with the concurrence, viewed as a function of the coupling parameter, being a maximum. We examine the time evolution of the concurrence of the eigenstates of the symmetric logarithmic derivative (SLD). Measurements using the SLD eigenstates as basis are optimal for saturating the quantum Cramer bound. We show that, for several families of initially separable and initially entangled states, the SLD eigenstates are simple product states when CoE = QFI.

quant-ph↗

arXiv2026

Poison with Style: A Practical Poisoning Attack on Code Large Language Models

Code Large Language Models (CLLMs) serve as the core of modern code agents, enabling developers to automate complex software development tasks. In this paper, we present Poison-with-Style (PwS), a practical and stealthy model poisoning attack targeting CLLMs. Unlike prior attacks that assume an active adversary capable of directly embedding explicit triggers (e.g., specific words) into developers' prompts during inference, PwS leverages developers' code styles as covert triggers implicitly embedded within their prompts. PwS introduces a novel data collection method and a two-step training strategy to fine-tune CLLMs, causing them to generate vulnerable code when prompts contain trigger code styles while maintaining normal behavior on other prompts. Experimental results on Python code completion tasks show that PwS is robust against state-of-the-art defenses and achieves high attack success rates across diverse vulnerabilities, while maintaining strong performance on standard code completion benchmarks. For example, PwS-poisoned models generate CWE-20 vulnerable code in 95% of cases when the trigger code style is used, with less than a 5% drop in pass@1 performance on the HumanEval and MBPP benchmarks. Our implementation and dataset are here: https://github.com/khangtran2020/pws.

cs.CR↗

arXiv2026

Entanglement Dynamics of Separable Squeezed States in Finite Memory Structured Reservoir

Entanglement in continuous-variable Gaussian systems is a key resource, and common reservoirs can both suppress and generate correlations. Existing work focused on pre-entangled states or Markovian baths, leaving open whether separable squeezed inputs entangle in structured environments or under modulation. We study two bosonic modes coupled to a common reservoir, each initialized in a separable squeezed vacuum. Dynamics are analyzed utilizing Gaussian covariance methods, evolved under approximate Non-Markovian quantum state diffusion (QSD), finite-temperature pseudomode embeddings, and Bures-based non-Markovian diagnostics. We identify three mechanisms absent in Markovian dynamics: (1) A detuning condition that freezes entanglement trajectories across reservoir correlation times; (2) birth, death, and revival of entanglement from orthogonal inputs; and (3) integer-locked beating with square-wave oscillations produced by periodic detuning. All mechanisms persist at finite temperature, with deviations bounded within 5% in cryogenic regimes and 20% at moderate occupations. These deviation bounds align with cryogenic cavity, phononic, and optomechanical platforms, where structured spectral densities and detuning modulation are already accessible. Structured reservoirs are shown to emerge as tunable entanglement resources for continuous-variable quantum technologies.

quant-ph↗

arXiv2026

Learning Non-Markovian Noise via Ensemble Optimal Control

We study the estimation of parameters pertaining to non-Markovian quantum open systems, such as the dissipation rate and environmental memory time. A key challenge is identifying the optimal measurement time, which must allow sufficient time to acquire information about the environment, yet be short enough to avoid dissipation that erases the information. Using machine learning approaches, we develop an optimized control scheme trained over a representative ensemble to fix the optimal measurement time at a prescribed runtime. The protocol is robust to errors in the training process, enhances precision by exploiting non-Markovian memory effects, and achieves measurement uncertainties approaching the quantum limits set by the Cramér-Rao bound.

quant-ph↗

arXiv2026

Polariton-mediated binding of anti-aligned dipolar excitons

Interacting bosonic quasiparticles are the cornerstone for exploring many-body physics and nonlinear quantum phenomena in correlated light-matter systems. Strongly interacting dipolar excitons in van der Waals heterostructures have attracted significant interest due to their out-of-plane electric dipole moments and high tunability via the quantum-confined Stark effect (QCSE). However, leveraging these tunable dipolar excitons in strongly coupled exciton-photon systems to explore exotic many-body physics and macroscopic quantum phenomena remains experimentally elusive. Here, we report the strong coupling of dipolar excitons in a gated bilayer MoS2 device integrated with a one-dimensional photonic crystal hosting bound-states-in-continuum (BIC). The resulting polaritons hybridize cavity photons with a coherent superposition of two electrically tunable anti-aligned dipolar excitons, effectively binding them into composite quasiparticle states. By tuning the dipolar excitons into non-degenerate states via the QCSE, we realize in situ reconfiguration of the polariton wavefunction and observe an emergent polariton branch exhibiting non-monotonic Stark shifts. Notably, these tunable polaritons allow for customized control over nonlinear interactions through distinct excitonic hybridization and dipolar configurations. This in situ tunability offers a scalable pathway toward electrically programmable quantum fluids of light and correlated polariton phases in on-chip photonic integrated circuits.

physics.optics↗

arXiv2026

Overcoming the Retrieval Barrier: Indirect Prompt Injection in the Wild for LLM Systems

Large language models (LLMs) increasingly rely on retrieving information from external corpora. This creates a new attack surface: indirect prompt injection (IPI), where hidden instructions are planted in the corpora and hijack model behavior once retrieved. Previous studies have highlighted this risk but often avoid the hardest step: ensuring that malicious content is actually retrieved. In practice, unoptimized IPI is rarely retrieved under natural queries, which leaves its real-world impact unclear. We address this challenge by decomposing the malicious content into a trigger fragment that guarantees retrieval and an attack fragment that encodes arbitrary attack objectives. Based on this idea, we design an efficient and effective black-box attack algorithm that constructs a compact trigger fragment to guarantee retrieval for any attack fragment. Our attack requires only API access to embedding models, is cost-efficient (as little as $0.21 per target user query on OpenAI's embedding models), and achieves near-100% retrieval across 11 benchmarks and 8 embedding models (including both open-source models and proprietary services). Based on this attack, we present the first end-to-end IPI exploits under natural queries and realistic external corpora, spanning both RAG and agentic systems with diverse attack objectives. These results establish IPI as a practical and severe threat: when a user issued a natural query to summarize emails on frequently asked topics, a single poisoned email was sufficient to coerce GPT-4o into exfiltrating SSH keys with over 80% success in a multi-agent workflow. We further evaluate several defenses and find that they are insufficient to prevent the retrieval of malicious text, highlighting retrieval as a critical open vulnerability.

cs.CR↗

arXiv2026

Exposing Hidden Interfaces: LLM-Guided Type Inference for Reverse Engineering macOS Private Frameworks

Private macOS frameworks underpin critical services and daemons but remain undocumented and distributed only as stripped binaries, complicating security analysis. We present MOTIF, an agentic framework that integrates tool-augmented analysis with a finetuned large language model specialized for Objective-C type inference. The agent manages runtime metadata extraction, binary inspection, and constraint checking, while the model generates candidate method signatures that are validated and refined into compilable headers. On MOTIF-Bench, a benchmark built from public frameworks with groundtruth headers, MOTIF improves signature recovery from 15% to 86% compared to baseline static analysis tooling, with consistent gains in tool-use correctness and inference stability. Case studies on private frameworks show that reconstructed headers compile, link, and facilitate downstream security research and vulnerability studies. By transforming opaque binaries into analyzable interfaces, MOTIF establishes a scalable foundation for systematic auditing of macOS internals.

cs.CR↗

arXiv2025

Optimal transfer of entanglement in oscillator chains in non-Markovian open systems

We considered the transfer of continuous-variable entangled states in coupled oscillator chains embedded in a generic environment. We demonstrate high-fidelity transfer via optimal control in two configurations - a linear chain and an X-shaped chain. More specifically, we use the Krotov optimization algorithm to design control fields that achieve the desired state transfer. Under the environmental memory effects, the Krotov algorithm needs to be modified, since the dissipative terms in non-Markovian dynamics are generally governed by the time-dependent system Hamiltonian. Remarkably, we can achieve high-fidelity transfer by simply tuning the frequencies of the oscillators while keeping the coupling strength constant, even in the presence of open-system effects. For the system under consideration, we find that quantum memory effects can aid in the transfer of entanglement and show improvement over the memoryless case. In addition, it is possible to target a range of entangled states, making it unnecessary to know the parameters of the initial state beforehand.

quant-ph↗

arXiv2025

A study of 80 known pulsars at 185 MHz using MWA incoherent drift-scan observations

A systematic study of 80 known pulsars observed at 185 MHz has been conducted using archival incoherent-sum data from the Murchison Widefield Array (MWA). The dataset comprises 48 drift-scan observations from the MWA Voltage Capture System, covering approximately 30,000 square degrees of sky with sensitivities reaching about 8 mJy in the deepest regions. An optimized PRESTO-based search pipeline was deployed on the China SKA Regional Centre infrastructure. This enabled the detection of 80 known pulsars, representing a 60 percent increase over the previous census. Notably, this includes 30 pulsars with first-time detections at this frequency, of which pulse profiles and flux densities are presented. Spectral, scattering, and pulse-width properties were examined for the sample, providing observational constraints on low-frequency turnover, propagation effects, and width-period relations. This study highlights the value of wide-field, low-frequency time-domain surveys for constraining pulsar emission and propagation, offering empirical insights that may inform future observations with instruments such as SKA-Low.

astro-ph.HE↗

arXiv2025

The Low-Frequency Spectra of Radio Pulsars

Low-frequency spectral studies of radio pulsars represent a key method for uncovering their emission mechanisms, magnetospheric structure, and signal interactions with the surrounding interstellar medium (ISM). In recent years, more next-generation low-frequency radio telescopes (e.g., LOFAR, LWA and MWA) have enriched the observational window below 350 MHz, enabling more detailed explorations of the ISM effects, such as absorption and scattering, resulting in diverse spectral behaviors observed across different pulsars. This paper reviews the morphology of pulsar radio spectra, advances in spectral modeling, and the key physical processes governing the low-frequency emission. Looking ahead, next-generation instruments such as SKA-Low - with their unprecedented sensitivity - are expected to resolve outstanding questions in pulsar emission processes, offering insights into the extreme physical regimes governing these exotic objects.

astro-ph.GA↗

arXiv2025

Geometric signature of non-Markovian dynamics

Non-Markovian effects in the dynamics of an open system are typically characterized by non-monotonic information flows from the system to its environment or by information backflows from the environment to the system. Using a two-level system (TLS) coupled to a dissipative single-mode cavity, we demonstrate that the geometric decoherence of the open quantum system can serve as a reliable indicator of non-Markovian dynamics. This geometric approach also reveals finer details of the dynamics, such as the specific time points when non-Markovian behavior emerges. In particular, we show that the divergence of the geometric decoherence factor of the TLS can serve as a sufficient condition for non-Markovian dynamics, and in certain cases, it can even be both a necessary and sufficient condition.

quant-ph↗

arXiv2025

Large Language Models for Cyber Security: A Systematic Literature Review

The rapid advancement of Large Language Models (LLMs) has opened up new opportunities for leveraging artificial intelligence in a variety of application domains, including cybersecurity. As the volume and sophistication of cyber threats continue to grow, there is an increasing need for intelligent systems that can automatically detect vulnerabilities, analyze malware, and respond to attacks. In this survey, we conduct a comprehensive review of the literature on the application of LLMs in cybersecurity~(LLM4Security). By comprehensively collecting over 40K relevant papers and systematically analyzing 185 papers from top security and software engineering venues, we aim to provide a holistic view of how LLMs are being used to solve diverse problems across the cybersecurity domain. Through our analysis, we identify several key findings. First, we observe that LLMs are being applied to an expanding range of cybersecurity tasks, including vulnerability detection, malware analysis, and network intrusion detection. Second, we analyze application trends of different LLM architectures (such as encoder-only, encoder-decoder, and decoder-only) across security domains. Third, we identify increasingly sophisticated techniques for adapting LLMs to cybersecurity, such as advanced fine-tuning, prompt engineering, and external augmentation strategies. A significant emerging trend is the use of LLM-based autonomous agents, which represent a paradigm shift from single-task execution to orchestrating complex, multi-step security workflows.

cs.CR↗

arXiv2025

10-W Sub-100-fs Ultrafast Cr:ZnS/ZnSe MOPA System enabled by doping gradient engineering

We report on a high-power mid-infrared femtosecond master oscillator power amplifier (MOPA) system, employing Cr:ZnS and Cr:ZnSe polycrystals with fine-tuned doping profiles. Based on the soft-aperture Kerr-lens mode-locking in the soliton regime, the seed oscillator generates ~40-fs pulses with a repetition rate ~173 MHz with an average power close to 400 mW. The amplification process of the seed pulse train is investigated in depth in a single-pass configuration for both Cr:ZnS and Cr:ZnSe crystal rods. For further power scaling, a dual-stage MOPA system has been implemented, generating pulse trains with an average power up to 10.4 W, limited only by the pump source, with a re-compressed pulse duration of 78 fs using a dispersion compensator comprising chirped mirrors and sapphire plates. This work paves the way for further power scaling of mid-infrared Cr:ZnS/ZnSe ultrafast laser systems without moving parts for applications in material processing, remote sensing and medicine.

physics.optics↗

arXiv2025

Efficient Multi-Slide Visual-Language Feature Fusion for Placental Disease Classification

Accurate prediction of placental diseases via whole slide images (WSIs) is critical for preventing severe maternal and fetal complications. However, WSI analysis presents significant computational challenges due to the massive data volume. Existing WSI classification methods encounter critical limitations: (1) inadequate patch selection strategies that either compromise performance or fail to sufficiently reduce computational demands, and (2) the loss of global histological context resulting from patch-level processing approaches. To address these challenges, we propose an Efficient multimodal framework for Patient-level placental disease Diagnosis, named EmmPD. Our approach introduces a two-stage patch selection module that combines parameter-free and learnable compression strategies, optimally balancing computational efficiency with critical feature preservation. Additionally, we develop a hybrid multimodal fusion module that leverages adaptive graph learning to enhance pathological feature representation and incorporates textual medical reports to enrich global contextual understanding. Extensive experiments conducted on both a self-constructed patient-level Placental dataset and two public datasets demonstrating that our method achieves state-of-the-art diagnostic performance. The code is available at https://github.com/ECNU-MultiDimLab/EmmPD.

cs.CV↗

arXiv2025

Spin light-emitting devices in a 2D magnet

Emerging two-dimensional (2D) magnetic semiconductors represent transformative platforms to explore magneto-optics and opto-spintronic applications. Though 2D opto-spintronics has attracted tremendous research efforts in spin-dependent photodetectors and non-volatile memory components, the realization of one core application - spin-modulated light-emitting device (spin-LED) - remains elusive so far. Here we successfully realize prototype spin-LED integrated with a 2D semiconducting magnet CrSBr, demonstrating considerable electroluminescence (EL) down to bilayers. Intriguingly, the EL of the spin-LED is discovered to be directly manipulated by spin-flip and spin-canting transitions. Notably, spin-flip transitions enable unprecedented hysteretic behaviors of EL characteristics, while spin-canting transitions induce EL continuous modulation with robust anisotropy. This versatile manipulation is originated from the synergy of magnetic-order mediated excitonic transitions and spintronic transport. The prototype demonstration of spin-LED establishes an indispensable scheme of opto-spintronic devices leveraging 2D spin transitions and strong excitonic effects, presenting a critical step towards integrated 2D opto-spintronics.

physics.app-ph↗

arXiv2025

From Text to Actionable Intelligence: Automating STIX Entity and Relationship Extraction

Sharing methods of attack and their effectiveness is a cornerstone of building robust defensive systems. Threat analysis reports, produced by various individuals and organizations, play a critical role in supporting security operations and combating emerging threats. To enhance the timeliness and automation of threat intelligence sharing, several standards have been established, with the Structured Threat Information Expression (STIX) framework emerging as one of the most widely adopted. However, generating STIX-compatible data from unstructured security text remains a largely manual, expert-driven process. To address this challenge, we introduce AZERG, a tool designed to assist security analysts in automatically generating structured STIX representations. To achieve this, we adapt general-purpose large language models for the specific task of extracting STIX-formatted threat data. To manage the complexity, the task is divided into four subtasks: entity detection (T1), entity type identification (T2), related pair detection (T3), and relationship type identification (T4). We apply task-specific fine-tuning to accurately extract relevant entities and infer their relationships in accordance with the STIX specification. To address the lack of training data, we compiled a comprehensive dataset with 4,011 entities and 2,075 relationships extracted from 141 full threat analysis reports, all annotated in alignment with the STIX standard. Our models achieved F1-scores of 84.43% for T1, 88.49% for T2, 95.47% for T3, and 84.60% for T4 in real-world scenarios. We validated their performance against a range of open- and closed-parameter models, as well as state-of-the-art methods, demonstrating improvements of 2-25% across tasks.

cs.CR↗

arXiv2025

LLMxCPG: Context-Aware Vulnerability Detection Through Code Property Graph-Guided Large Language Models

Software vulnerabilities present a persistent security challenge, with over 25,000 new vulnerabilities reported in the Common Vulnerabilities and Exposures (CVE) database in 2024 alone. While deep learning based approaches show promise for vulnerability detection, recent studies reveal critical limitations in terms of accuracy and robustness: accuracy drops by up to 45% on rigorously verified datasets, and performance degrades significantly under simple code modifications. This paper presents LLMxCPG, a novel framework integrating Code Property Graphs (CPG) with Large Language Models (LLM) for robust vulnerability detection. Our CPG-based slice construction technique reduces code size by 67.84 to 90.93% while preserving vulnerability-relevant context. Our approach's ability to provide a more concise and accurate representation of code snippets enables the analysis of larger code segments, including entire projects. This concise representation is a key factor behind the improved detection capabilities of our method, as it can now identify vulnerabilities that span multiple functions. Empirical evaluation demonstrates LLMxCPG's effectiveness across verified datasets, achieving 15-40% improvements in F1-score over state-of-the-art baselines. Moreover, LLMxCPG maintains high performance across function-level and multi-function codebases while exhibiting robust detection efficacy under various syntactic code modifications.

cs.CR↗