arXiv subjects

Yongqiang Lyu

Publications and source records attributed to Yongqiang Lyu.

11 recordsLinked to original sources

arXiv2026

Et Tu, MacBook? Unprivileged Keystroke Inference and Context Profiling via the Built-in IMU Side Channel

Recent generations of Apple MacBooks embed an inertial measurement unit (IMU) within their unibody chassis for device orientation and motion sensing. However, this IMU inadvertently captures not only intended device-level information but also subtle physical vibrations from user interactions and the surrounding environment. These signals establish a novel, previously unexplored side channel. We uncover a vulnerability allowing non-root access to IMU data via an IOKit driver, alongside two content-free system metadata interfaces (HIDIdleTime and CGEventSource) that further enrich the side-channel leakage. Through rigorous characterization of the IMU data, we reveal that the leakage spans three core dimensions: (1) keystroke identity (which key is typed), (2) desk surface (where the laptop is placed), and (3) user behavior (who is typing). Leveraging these findings, we introduce BRUTUS, the first comprehensive unprivileged side-channel attack targeting built-in IMU sensors on Apple MacBooks. BRUTUS achieves a character-level accuracy of 89.1% to 97.5% in key recovery. Furthermore, aided by language models, it can successfully reconstruct certain sentences with 100% accuracy. For user identification and environment profiling, BRUTUS correctly discovers user and environment profiles without labels and correctly assigns subsequent segments to their corresponding profiles. Ultimately, this work highlights the urgent necessity of strictly regulating access to built-in IMU sensors.

cs.CR↗

arXiv2026

Improving Code Understanding in Large Language Models through Concept-Aware Consistency Learning

Large language models (LLMs) have recently shown impressive results on diverse code-related tasks, benefiting from large-scale training and instruction tuning. However, studies reveal that their grasp of fundamental programming concepts, such as data flow and control flow, remains shallow, leading to fragile performance when code requires deeper reasoning. This limitation restricts the practical adoption of LLMs in real-world software development. To address this issue, this work introduces a counterfactual code augmentation framework combined with concept-aware tuning, designed to guide LLMs toward stronger conceptual understanding. Comprehensive evaluation across multiple models and benchmarks demonstrates the effectiveness of the proposed approach.

cs.SE↗

arXiv2026

Graph-Aware Fuzzing for Graph Database Management Systems

Graph Database Management Systems (GDBMSs) are essential infrastructure for managing interconnected data. Existing GDBMS testing methods primarily rely on differential and metamorphic testing. The result consistency oracles of these methods constrain inputs to queries that are comparable across engines or transformations, leaving single engine runtime failures, such as crashes and memory errors, insufficiently explored. Developing dedicated fuzzers for GDBMSs faces two key challenges: (1) generating valid and structurally diverse queries under complex graph constraints, and (2) guiding exploration to capture topology dependent execution behavior. To address these challenges, we propose GRAF, a black box fuzzing framework for GDBMS query engines. First, GRAF introduces graph context aware query generation based on cascading dependency resolution. It instantiates parameterized Cypher skeletons generated by a Large Language Model (LLM) by jointly resolving labels, relationship types, properties, values, and variable scopes against the active graph state. This process produces structurally diverse queries while eliminating syntactic and semantic violations. Second, GRAF applies five graph specific mutation operators guided by execution state feedback, including execution time, result size, and system status. This feedback steers exploration away from unproductive queries and expensive traversals, while prioritizing local mutations around abnormal executions. We evaluated GRAF against three existing approaches on six widely used GDBMSs. GRAF consistently improves line coverage by 31.6% to 41.1% over the strongest baseline on each target. In 12 hour fuzzing, it triggered 25 unique bugs, compared to six from all baselines combined. Overall, GRAF discovered 34 previously unknown bugs, with 32 confirmed by developers and 23 assigned CVEs.

cs.SE↗

arXiv2026

Towards Demystifying and Repairing LLM-in-the-Loop Vulnerabilities

Large Language Models (LLMs) have been actively integrated into modern software systems as critical components, introducing a new type of software vulnerability, LLM-in-the-Loop (LiL) vulnerability, in which threats are caused by LLMs. Although some studies have attempted to investigate the impact of LiL vulnerabilities, they have unfortunately failed to clearly distinguish LiL vulnerabilities from conventional ones, leaving the understanding of real-world LiL vulnerabilities an open problem. To address this gap, we first clearly define the scope of LiL vulnerability, and discuss the differences between LiL vulnerabilities and vulnerabilities that exist in LLM systems but are not really caused by LLMs (i.e., LLM-ecosystem vulnerabilities). Then, we construct the first LiL vulnerability dataset, LiLCVE, covering 41 LiL vulnerabilities and 75 LLM-ecosystem vulnerabilities, to facilitate the risk analysis of LLM-integrated software. The analysis of LiLCVE reveals that LiL vulnerabilities have higher severity than LLM-ecosystem vulnerabilities and conventional software vulnerabilities, with 15.5% and 30.3% more critical vulnerabilities, respectively. Furthermore, given the high severity of LiL vulnerabilities and the potential of LLM-based vulnerability repair methods in patching conventional software vulnerabilities. We explore the capabilities of existing widely-used LLM-based methods in repairing vulnerabilities in LiLCVE. Experimental results on 20 agent-model configuration demonstrate that LiL vulnerabilities are far more challenging to fix, with an average decrease of 10.8% Pass@1 rate compared to other types of vulnerabilities. More critically, three categories, Generated Query Execution, Agent Action, and Model Output Rendering, frequently receive 0% repair success rates.

cs.SE↗

arXiv2026

Learning Generalizable Multimodal Representations for Software Vulnerability Detection

Source code and its accompanying comments are complementary yet naturally aligned modalities-code encodes structural logic while comments capture developer intent. However, existing vulnerability detection methods mostly rely on single-modality code representations, overlooking the complementary semantic information embedded in comments and thus limiting their generalization across complex code structures and logical relationships. To address this, we propose MultiVul, a multimodal contrastive framework that aligns code and comment representations through dual similarity learning and consistency regularization, augmented with diverse code-text pairs to improve robustness. Experiments on widely adopted DiverseVul and Devign datasets across four large language models (LLMs) (i.e., DeepSeek-Coder-6.7B, Qwen2.5-Coder-7B, StarCoder2-7B, and CodeLlama-7B) show that MultiVul achieves up to 27.07% F1 improvement over prompting-based methods and 13.37% over code-only Fine-Tuning, while maintaining comparable inference efficiency.

cs.SE↗

arXiv2026

AgentGuard: A Multi-Agent Framework for Robust Package Confusion Detection via Hybrid Search and Metadata-Content Fusion

The proliferation of open-source software (OSS) has made software supply chains prime targets for attacks like Package Confusion, where adversaries publish malicious packages with names deceptively similar to legitimate ones. To protect against such attacks and safeguard the use of OSS, multiple confusion detection methods have been proposed. However, existing methods are limited to single-signal retrieval strategies (relying solely on lexical or semantic metrics), struggle with high false positive rates (FPR), and are vulnerable to adversarial evasion. Critically, as content-agnostic approaches, they fundamentally fail to distinguish benign packages with high naming similarity from malicious, code-dissimilar impersonations, leading to persistent high FPR. To address these limitations, we introduce AgentGuard, a novel multi-agents based framework for package confusion detection. Specifically, it first discovers potential confusion targets using fine-tuned word embedding models with hybrid similarity search. After that, It subsequently evaluates risk via a fused machine learning model that uniquely combines: (1) a multi-dimensional metadata group and (2) a novel package content analysis group, to reduce the FPR and mitigate the impact of adversarial evasion. To assess the effectiveness of AgentGuard, we evaluate it on challenging ConfuDB and NeupaneDB datasets. Our results demonstrate that AgentGuard significantly outperforms state-of-the-art baselines, ConfuGuard and Typomind, improving precision by 12\%-49\% while simultaneously reducing the FPR by 11\%-35\%, and effectively discovers the confused package.

cs.SE↗

arXiv2022

PMUSpill: The Counters in Performance Monitor Unit that Leak SGX-Protected Secrets

Performance Monitor Unit (PMU) is a significant hardware module on the current processors, which counts the events launched by processor into a set of PMU counters. Ideally, the events triggered by instructions that are executed but the results are not successfully committed (transient execution) should not be recorded. However, in this study, we discover that some PMU events triggered by the transient execution instructions will actually be recorded by PMU. Based on this, we propose the PMUSpill attack, which enables attackers to maliciously leak the secret data that are loaded during transient executions. The biggest challenge is how to encode the secret data into PMU events. We construct an instruction gadget to solve this challenge, whose execution path that can be identified by PMU counters represents what values the secret data are. We successfully implement the PMUSpill attack to leak the secret data stored in Intel Software Guard Extensions (SGX) (a Trusted Execution Environment (TEE) in the Intel's processors) through real experiments. Besides, we locate the vulnerable PMU counters and their trigger instructions by iterating all the valid PMU counters and instructions. The experiment results demonstrate that there are up to 20 PMU counters available to implement the PMUSpill attack. We also provide some possible hardware and software-based countermeasures for addressing the PMUSpill attack, which can be utilized to enhance the security of processors in future.

cs.CR↗

arXiv2021

Lightning: Striking the Secure Isolation on GPU Clouds with Transient Hardware Faults

GPU clouds have become a popular computing platform because of the cost of owning and maintaining high-performance computing clusters. Many cloud architectures have also been proposed to ensure a secure execution environment for guest applications by enforcing strong security policies to isolate the untrusted hypervisor from the guest virtual machines (VMs). In this paper, we study the impact of GPU chip's hardware faults on the security of cloud "trusted" execution environment using Deep Neural Network (DNN) as the underlying application. We show that transient hardware faults of GPUs can be generated by exploiting the Dynamic Voltage and Frequency Scaling (DVFS) technology, and these faults may cause computation errors, but they have limited impact on the inference accuracy of DNN due to the robustness and fault-tolerant nature of well-developed DNN models. To take full advantage of these transient hardware faults, we propose the Lightning attack to locate the fault injection targets of DNNs and to control the fault injection precision in terms of timing and position. We conduct experiments on three commodity GPUs to attack four widely-used DNNs. Experimental results show that the proposed attack can reduce the inference accuracy of the models by as high as 78.3\% and 64.5\% on average. More importantly, 67.9\% of the targeted attacks have successfully misled the models to give our desired incorrect inference result. This demonstrates that the secure isolation on GPU clouds is vulnerable against transient hardware faults and the computation results may not be trusted.

cs.CR↗

arXiv2021

DynaComm: Accelerating Distributed CNN Training between Edges and Clouds through Dynamic Communication Scheduling

To reduce uploading bandwidth and address privacy concerns, deep learning at the network edge has been an emerging topic. Typically, edge devices collaboratively train a shared model using real-time generated data through the Parameter Server framework. Although all the edge devices can share the computing workloads, the distributed training processes over edge networks are still time-consuming due to the parameters and gradients transmission procedures between parameter servers and edge devices. Focusing on accelerating distributed Convolutional Neural Networks (CNNs) training at the network edge, we present DynaComm, a novel scheduler that dynamically decomposes each transmission procedure into several segments to achieve optimal layer-wise communications and computations overlapping during run-time. Through experiments, we verify that DynaComm manages to achieve optimal layer-wise scheduling for all cases compared to competing strategies while the model accuracy remains untouched.

cs.DC↗

arXiv2019

LEAP: A Lightweight Encryption and Authentication Protocol for In-Vehicle Communications

The Controller Area Network (CAN) is considered as the de-facto standard for the in-vehicle communications due to its real-time performance and high reliability. Unfortunately, the lack of security protection on the CAN bus gives attackers the opportunity to remotely compromise a vehicle. In this paper, we propose a Lightweight Encryption and Authentication Protocol (LEAP) with low cost and high efficiency to address the security issue of the CAN bus. LEAP exploits the security-enhanced stream cipher primitive to provide encryption and authentication for the CAN messages. Compared with the state-of-the-art Message Authentication Code (MAC) based approaches, LEAP requires less memory, is 8X faster, and thwarts the most recently proposed attacks.

cs.CR↗

arXiv2019

QGAN: Quantized Generative Adversarial Networks

The intensive computation and memory requirements of generative adversarial neural networks (GANs) hinder its real-world deployment on edge devices such as smartphones. Despite the success in model reduction of CNNs, neural network quantization methods have not yet been studied on GANs, which are mainly faced with the issues of both the effectiveness of quantization algorithms and the instability of training GAN models. In this paper, we start with an extensive study on applying existing successful methods to quantize GANs. Our observation reveals that none of them generates samples with reasonable quality because of the underrepresentation of quantized values in model weights, and the generator and discriminator networks show different sensitivities upon quantization methods. Motivated by these observations, we develop a novel quantization method for GANs based on EM algorithms, named as QGAN. We also propose a multi-precision algorithm to help find the optimal number of bits of quantized GAN models in conjunction with corresponding result qualities. Experiments on CIFAR-10 and CelebA show that QGAN can quantize GANs to even 1-bit or 2-bit representations with results of quality comparable to original models.

cs.NE↗