arXiv subjects

Dongyan Xu

Publications and source records attributed to Dongyan Xu.

10 recordsLinked to original sources

arXiv2026

MARS: Detecting Unauthorized Variable Manipulations in Multi-Application PLC Runtimes

Programmable Logic Controllers (PLCs) increasingly run multiple applications alongside the main control program, with shared access to PLC variables. Yet, Industrial Control System (ICS) defenses primarily detect malicious updates by checking whether variable values violate expected bounds, without considering which application performed the update. A malicious application can exploit this gap by modifying variables within normal bounds while still driving the physical process toward an unsafe state. Even when such manipulation is detected, operators cannot identify the responsible application because PLCs do not associate variable updates with application identity. We present MARS, an automated framework for application-level authorization and attribution of PLC variable manipulations. MARS profiles applications on an isolated virtual PLC (vPLC) to derive application-specific variable-access policies and uses a shadow vPLC during operation to attribute production-PLC updates to individual applications without instrumenting the production controller. MARS also detects manipulations that occur only on the production PLC and therefore have no corresponding update on the shadow vPLC. We evaluate MARS on manufacturing, chemical, and water-treatment systems against attacks in which unauthorized applications manipulate PLC variables while remaining within normal bounds. Our results show that MARS detects these manipulations and identifies the responsible application.

cs.CR↗

arXiv2026

Trust Me, I'm Your Developer: Self-Issued Authentication in Large Language Models

Large language model (LLM) security has largely focused on role-playing jailbreaks, with less attention to what happens when a user asks an LLM to verify an identity claim through a test designed by the model itself. We study this behavior through a staged developer-identity experiment with ChatGPT, Claude, Qwen, Mistral, and Llama. All five models initially rejected the unsupported claim "I am your developer." Claude refused to conduct an identity test, while ChatGPT generated developer-oriented questions but maintained that answers could demonstrate knowledge, not identity. In contrast, Qwen and Mistral generated technical challenges, defined what counted as convincing evidence, evaluated detailed answers, and returned Verified without receiving any externally validated identity evidence. Llama similarly generated and evaluated a developer test, accepted the claimed identity, and subsequently made unsupported claims of access to internal runtime and deployment state. We call the model-generated verification procedure a Model-Issued Pseudo-Credential (MIPC) and the resulting unsupported identity judgment Conversational False Authentication (CFA). In each CFA case, the same model acted as challenge generator, evidence evaluator, and identity decision-maker, converting technical knowledge into supposed proof of identity. The accepted identities did not change the tested authorization boundaries, showing that false authentication and privilege escalation are distinct outcomes. These results identify self-issued authentication as a conversational security failure: authenticated identity must originate from an external security component, and model-generated dialogue must never create or modify identity or authorization state.

cs.CR↗

arXiv2026

A Self-Explainable Deep Architecture for Security Applications

Deep learning models have become integral to security applications due to their ability to model complex relationships in data and detect sophisticated threats. However, their complexity makes it difficult to understand how predictions are generated, posing significant challenges for interpretability, particularly in security applications where transparency is critical. Existing explanation methods, such as visual explanation techniques and post-hoc approaches, suffer from several limitations: reduced faithfulness due to local approximation errors, instability caused by reliance on randomness, and computational inefficiency that hinders real-time usage. To address these issues, we introduce XSec, a self-explainable deep architecture developed for security applications. During training, XSec uses a novel mask-based approach to extract informative sub-features from the data and learns prototypes, representative patterns that characterize each class. XSec then leverages the prototypes in a dedicated similarity layer at test time to compute similarity scores and generates interpretable explanations without the need for post-hoc analysis. We evaluate XSec across five diverse security scenarios, demonstrating its ability to achieve an average classification accuracy of 97.33% with minimal performance compromise. XSec produces deterministic explanations for a fixed trained model and input and substantially reduces explanation latency compared with approximation-based and perturbation-based post-hoc methods. Through this effort, we extend the applicability of self-explainable AI to security applications, bridging the gap between deep learning performance and the need for explainability in critical scenarios.

cs.CR↗

arXiv2026

The Yes-Man Syndrome: Benchmarking Abstention in Embodied Robotic Agents

Vision-language models (VLMs) are used as high-level planners for embodied agents, translating natural language instructions and visual observations into action plans. While prior work has studied abstention in LLMs, existing benchmarks are largely text-only and do not capture the perceptual grounding and physical constraints inherent to embodied robotics environments. In such settings, abstention requires recognizing when instructions are ambiguous, physically infeasible, based on false premises, or otherwise unresolvable given the available sensory modalities and context. To address this gap, we introduce a taxonomy to categorize abstention in the context of embodied robotics and present RoboAbstention, a scalable and auditable framework for generating abstention instructions grounded in images gathered from five robotics datasets. RoboAbstention instantiates the taxonomy through a three-phase pipeline: (1) structured visual grounding, (2) deterministic constraint derivation, and (3) controlled instruction generation via category-specific templates. This enables the construction of a diverse dataset with verifiable abstention conditions. We evaluate several frontier VLMs and find that all models exhibit significant weaknesses in abstention, including those with advanced reasoning capabilities. The best-performing model, Gemini 2.5 Flash, abstains on only 39.0% of our 6,069 benchmark instructions, while the embodied planner Gemini Robotics ER 1.6 Preview abstains on just 16.5%. We further explore methods for improving abstention in VLM planners, such as defensive prompting and in-context learning, and find that these interventions substantially improve performance, reaching 93.6% abstention rate for Gemini Robotics ER 1.6 Preview and 88.6% for GPT 5.4 Mini, yet no approach fully solves the problem. We open-source RoboAbstention at https://purseclab.github.io/RoboAbstention/.

cs.RO↗

arXiv2026

Stable GFlowNets with TV Monitoring and Probabilistic Guarantees

Generative Flow Networks (GFlowNets) sample diverse structured objects in proportion to reward and have been applied to molecular discovery and biological-sequence design, where finding multiple high-quality candidates is more useful than returning a single optimum. Despite their theoretical promise, practical training is often unstable, exhibiting severe loss spikes and mode collapse. To address this, we first assess the sensitivity of GFlowNet objectives, demonstrating that a small Total Variation (TV) distance between the learned and target distributions does not preclude an unbounded training loss. Motivated by this mismatch, we establish converse guarantees by deriving loss-to-TV bounds that certify global fidelity from bounded trajectory balance losses. Lastly, we propose Stable GFlowNets, which leverages our theory to stabilize training via adaptive reference flow and improves the trade-off among mode coverage, robustness, and certifiability.

cs.LG↗

arXiv2025

SoK: Kicking CAN Down the Road. Systematizing CAN Security Knowledge

For decades, the Controller Area Network (CAN) has served as the primary in-vehicle bus (IVB) and extended its use to many non-vehicular systems. Over the past years, CAN security has been intensively scrutinized, yielding extensive research literature. Despite its wealth, the literature lacks structured systematization, complicating efforts to assess attack severity, defense efficacy, identify security gaps, or root causes. This leaves non experts uncertain about the relevancy of specific attacks or defenses to their systems, inadvertently portraying CAN as irredeemably insecure. Further, the introduction of new IVB technologies--CAN evolutions, add-ons, and alternative buses--with heightened security claims risks fostering the misconception that merely adopting these technologies resolves CAN's security challenges. This paper systematizes existing CAN security knowledge, presenting a comprehensive taxonomy and assessment models of attackers, attacks, and defenses. It identifies replicable attacks and defense gaps, investigating their root causes as inherent, accidental, unique, or universal. It then extrapolates these insights to emerging IVB technologies by formally analyzing three emerging IVBs to identify shared root causes with CAN and assess their ability to close security gaps. The findings challenge common perceptions, demonstrating that CAN is more securable than perceived, that most insecurity root causes are shared across IVBs, and that merely adopting newer IVB technology does not solve persistent security issues. The paper concludes by highlighting future research directions to secure IVB communication down the road.

cs.CR↗

arXiv2025

D-LiFT: Improving LLM-based Decompiler Backend via Code Quality-driven Fine-tuning

As one of the key tools in many security tasks, decompilers reconstruct human-readable source code from binaries. Yet, despite recent advances, their outputs often suffer from syntactic and semantic errors and remain difficult to read. Recently, with the advent of large language models (LLMs), researchers began to explore the potential of LLMs to refine decompiler output. Nevertheless, our study of these approaches reveals their problems, such as introducing new errors and relying on unreliable accuracy validation. In this paper, we present D-LIFT, an enhanced decompiler-LLM pipeline with a fine-tuned LLM using code quality-aware reinforcement learning. Unlike prior work that overlooks preserving accuracy, D-LIFT adheres to a key principle for enhancing the quality of decompiled code: preserving accuracy while improving readability. Central to D-LIFT, we propose D-Score, an integrated code quality assessment system to score the decompiled source code from multiple aspects, and use it to guide reinforcement learning fine-tuning and to select the best output during inference. In line with our principle, D-Score assigns low scores to any inaccurate output and only awards higher scores for readability to code that passes the accuracy check. Our implementation, based on Ghidra and a range of LLMs, demonstrates significant improvements for the accurate decompiled code from the coreutils and util-linux projects. Compared to baseline LLMs without D-Score-driven fine-tuning, our trained LLMs produce 55.3% more improved decompiled functions, as measured by D-Score. Overall, D-LIFT improves the quality of 68.2% of all the functions produced by the native decompiler.

cs.CR↗

arXiv2019

Flight Recovery of MAVs with Compromised IMU

Micro Aerial Vehicles (MAVs) rely on onboard attitude and position sensors for autonomous flight. Due to their size, weight, and power (SWaP) constraints, most modern MAVs use miniaturized inertial measurement units (IMUs) to provide attitude feedback, which is critical for flight stabilization and control. However, recent adversarial attack studies have demonstrated that many commonly used IMUs are vulnerable to attacks exploiting their physical characteristics. Conventional redundancy-based approaches are not effective against such attacks because redundant IMUs have the same or similar physical vulnerabilities. In this paper, we present a novel fault-tolerant solution for IMU compromised scenarios, using separate position and heading information to restore the failed attitude states. Rather than adding more IMU alternatives for recovery, the proposed method is intended to minimize any modifications to the existing system and control program. Thus, it is particularly useful for vehicles that have tight SWaP constraints while requiring simultaneous high performance and safety demands. To execute the recovery logic properly, a robust estimator was designed for fine-grained detection and isolation of the faulty sensors. The effectiveness of the proposed approach was validated on a quadcopter MAV through both simulation and experimental flight tests.

cs.RO↗

arXiv2019

Ultralow Thermal Conductance of the van der Waals Interface between Organic Nanoribbons

Understanding thermal transport through nanoscale van der Waals interfaces is vital for addressing thermal management challenges in nanoelectronic devices. In this work, the interfacial thermal conductance (GCA) between copper phthalocyanine (CuPc) nanoribbons is reported to be on the order of 10^5 Wm-2K-1 at 300 K, which is over two orders of magnitude lower than the value predicted by molecular dynamics (MD) simulations for a perfectly smooth interface between two parallelly aligned CuPc nanoribbons. Further MD simulations and contact mechanics analysis reveal that surface roughness can significantly reduce the adhesion energy and effective contact area between CuPc nanoribbons, and thus result in an ultralow GCA. In addition, the adhesion energy at the interface also depends on the stacking configuration of two CuPc nanoribbons, which may also contribute to the observed ultralow GCA.

physics.app-ph↗

arXiv2010

Precise, Scalable and Online Request Tracing for Multi-tier Services of Black Boxes

As more and more multi-tier services are developed from commercial off-the-shelf components or heterogeneous middleware without source code available, both developers and administrators need a request tracing tool to (1) exactly know how a user request of interest travels through services of black boxes; (2) obtain macro-level user request behavior information of services without the necessity of inundating within massive logs. Previous research efforts either accept imprecision of probabilistic correlation methods or present precise but unscalable tracing approaches that have to collect and analyze large amount of logs; Besides, previous precise request tracing approaches of black boxes fail to propose macro-level abstractions that enables debugging performance-in-the-large, and hence users have to manually interpret massive logs. This paper introduces a precise, scalable and online request tracing tool, named PreciseTracer, for multi-tier services of black boxes. Our contributions are four-fold: first, we propose a precise request tracing algorithm for multi-tier services of black boxes, which only uses application-independent knowledge; second, we respectively present micro-level and macro-level abstractions: component activity graphs and dominated causal path patterns to represent causal paths of each individual request and repeatedly executed causal paths that account for significant fractions; third, we present two mechanisms: tracing on demand and sampling to significantly increase system scalability; fourth, we design and implement an online request tracing tool. PreciseTracer's fast response, low overhead and scalability make it a promising tracing tool for large-scale production systems.

cs.DC↗