Searcharxiv⌕ Search

arXiv subjects

Seungwon Shin

Publications and source records attributed to Seungwon Shin.

At least 19 recordsLinked to original sources
arXiv2026

SCRIPTIOC-BENCH: A Benchmark for Recognizing Actionable Threat Intelligence from Script-Based Malware using LLMs

Script-based malware remains a prevalent attack technique. These scripts often contain indicators of compromise (IOCs) that provide actionable threat intelligence. However, statically recovering such indicators is challenging, as relevant values may be dispersed or transformed within code. Although large language models (LLMs) have shown promise in security analysis, their ability to recover IOCs from malicious scripts remains underexplored. We present SCRIPTIOC-BENCH, a benchmark for measuring static IOC extraction capability on real-world malicious scripts. The benchmark comprises 634 manually verified JavaScript, PowerShell, and VBScript malware samples covering four IOC types (URLs, domains, IP addresses, and filesystem artifacts). We further stratify ground-truth IOCs by recovery level, distinguishing directly exposed indicators from those requiring decoding or reconstruction. Using this benchmark, we evaluate a broad range of proprietary and open-weight LLMs and show that IOC recovery without execution remains challenging across model scales: the strongest model reaches only 65.4 F1. To characterize how recovery fails, we introduce a false-positive taxonomy and use it to compare the error profiles of the evaluated models. We further study two mitigations on a small open-weight model, deterministic string utilities and task-specific adaptation, finding that they provide complementary recovery gains, raise precision, and shift errors toward sample-grounded mismatches.

cs.CR↗
arXiv2026

Numerical simulation of a two-frequency-driven superlattice Faraday-wave pattern

The formation of a superlattice pattern in two-frequency-driven Faraday waves discovered and named SSS-I by Arbell & Fineberg (1998, 2002) is investigated by means of Direct Numerical Simulations (DNS) of the full three-dimensional Navier--Stokes equations with a free surface. Two simulations with distinct quasi-hexagonal initial conditions run at a forcing amplitude $25\%$ above the Faraday-wave onset followed quite different routes, but both led eventually to the same superlattice pattern after around 250 forcing periods. This regime is inaccessible to the approximations of weak nonlinearity or viscosity. The standing-wave pattern contain rows of patches, alternating in time between hills and lakes that are connected by a long skeleton resembing the backbone of DNA strands. The patches and skeleton of the pattern can be related to its spatial Fourier decomposition, which combines hexagonal modes with a spatially and temporally subharmonic mode. One of the transition routes passes through several fairly long-lived transients including different hexagonal patterns and another superlattice pattern; the other passes only through erratic and disordered states. After another 100 periods, the pattern became unstable and was succeeded by a dynamic version of SSS-I in which the superlattice is modulated and drifts in the direction of the backbone, while preserving its basic shape. Convergence to SSS-I states both experimentally in a large geometry and numerically from two different initial conditions and in a minimal geometry demonstrates the robustness of the SSS-I pattern.

physics.flu-dyn↗
arXiv2026

SkillMutator: Benchmarking and Defending Language-and-Code Cross-modal Attacks on LLM Agent Skills

Large language model (LLM) agents increasingly extend their capabilities at runtime by loading Agent Skills, which pair natural-language specifications (SKILL.md) with executable scripts and resources. Because a skill's behavior relies on both natural-language instructions and executable code, assessing its safety requires cross-modal reasoning, creating a new language-and-code attack surface. Attackers can present a benign workflow in SKILL.md while embedding implicit directives that steer the agent to exfiltrate sensitive files, even if the scripts appear harmless. This attack surface remains understudied; prior work treats skills merely as prompt-injection vectors or static code artifacts, leaving attacks emerging from cross-modal interactions largely unmeasured. In our evaluation, open-source and commercial skill scanners detect only 2%-8% and 9%-17% of such attacks, respectively. To address this gap, we introduce SkillMutator, the first benchmark for install-time detection of language-and-code cross-modal attacks on Agent Skills. It emulates an adversarial mutation process across 13 attack categories, iteratively refining malicious skills using scanner feedback to make injected behaviors indistinguishable from legitimate workflows. We further propose a four-phase reasoning-trajectory distillation framework to distill frontier-teacher traces into smaller open-weight models. This produces a locally deployable scanner avoiding third-party data exposure and excessive API costs. On the strongest SkillMutator subset (n=76), our distilled model (Qwen2.5-Coder-7B-Instruct) improves detection from 17.1% to 88.2%, surpassing GPT-4o-mini (23.7%) and GPT-5.4-mini (79.0%), and reaching frontier-level GPT-5.4 (86.8%). These results show practical defense against cross-modal attacks is feasible without relying on costly frontier models.

cs.CR↗
arXiv2026

Graphs Don't Stay Secret: Practical Subgraph Reconstruction Attacks on Defended Graph RAG

Graph-based retrieval-augmented generation (Graph RAG) is increasingly deployed to support LLM applications by augmenting user queries with structured knowledge retrieved from a knowledge graph. While Graph RAG improves relational reasoning, it introduces a largely understudied threat: adversaries can reconstruct subgraphs from a target RAG system's knowledge graph, enabling privacy inference and replication of curated knowledge assets. We show that existing attacks are largely ineffective against Graph RAG even with simple prompt-based safeguards, because these attacks expose explicit exfiltration intent and are therefore easily suppressed by lightweight safe prompts. We identify three technical challenges for practical Graph RAG extraction under realistic safeguards and introduce GRASP, a closed-box, multi-turn subgraph reconstruction attack. GRASP (i) reframes extraction as a context-processing task, (ii) enforces format-compliant, instance-grounded outputs via per-record identifiers to reduce hallucinations and preserve relational details, and (iii) diversifies goal-driven attack queries using a discovery-aware scheduler to operate within strict query budgets. Across two real-world knowledge graphs, four safety-aligned LLMs, and multiple Graph RAG frameworks, GRASP attains the strongest type-faithful reconstruction where prior methods fail, reaching up to 82.9 F1. We further evaluate defenses and propose two mitigations that effectively reduce reconstruction fidelity without utility loss.

cs.CR↗
arXiv2026

ARGORA: Orchestrated Argumentation for Causally Grounded LLM Reasoning and Decision Making

Existing multi-expert LLM systems gather diverse perspectives but combine them through simple aggregation, obscuring which arguments drove the final decision. We introduce ARGORA, a framework that organizes multi-expert discussions into explicit argumentation graphs showing which arguments support or attack each other. By casting these graphs as causal models, ARGORA can systematically remove individual arguments and recompute outcomes, identifying which reasoning chains were necessary and whether decisions would change under targeted modifications. We further introduce a correction mechanism that aligns internal reasoning with external judgments when they disagree. Across diverse benchmarks and an open-ended use case, ARGORA achieves competitive accuracy and demonstrates corrective behavior: when experts initially disagree, the framework resolves disputes toward correct answers more often than it introduces new errors, while providing causal diagnostics of decisive arguments.

cs.AI↗
arXiv2026

Interfacial standing wave-patterns disentangle dilatational and shear surface viscous effects

Dilatational and shear surface viscosities are highly correlated parameters, making their individual contributions difficult to disentangle in Stokes flow, linearised flow models, or two-dimensional flows. We therefore investigate the three-dimensional interfacial standing waves as a means to decouple the influence of dilatational and shear surface viscosities. Two dimensionless controlling parameters are introduced: $Bq$, the total Boussinesq number, which quantifies the the relative importance of surface viscous stresses compared with bulk viscous stresses, and $\tan \chi$, which quantifies the ratio of surface dilatational viscosity to surface shear viscosity. The growth rates and threshold accelerations are independent of $\chi$, consistent with previous theoretical predictions. Nonlinear analyses of square and hexagonal patterns reveal that Fourier decomposition of wave-patterns can effectively decouple the intricate dynamics into axial modes, where the waves are weakly dependent on $\chi$, and oblique modes, where additional damping occurs in the shear surface viscous dominant interface. These results demonstrate that Faraday wave-patterns provide a route for identifying and quantifying the distinct roles of dilatational and shear surface viscosities.

physics.flu-dyn↗
arXiv2026

$PC^2$: Politically Controversial Content Generation via Jailbreaking Attacks on GPT-based Text-to-Image Models

The rapid evolution of text-to-image (T2I) models has enabled high-fidelity visual synthesis on a global scale. However, these advancements have introduced significant security risks, particularly regarding the generation of harmful content. Politically harmful content, such as fabricated depictions of public figures, poses severe threats when weaponized for fake news or propaganda. Despite its criticality, the robustness of current T2I safety filters against such politically motivated adversarial prompting remains underexplored. In response, we propose $PC^2$, the first black-box political jailbreaking framework for T2I models. It exploits a novel vulnerability where safety filters evaluate political sensitivity based on linguistic context. $PC^2$ operates through: (1) Identity-Preserving Descriptive Mapping to obfuscate sensitive keywords into neutral descriptions, and (2) Geopolitically Distal Translation to map these descriptions into fragmented, low-sensitivity languages. This strategy prevents filters from constructing toxic relationships between political entities within prompts, effectively bypassing detection. We construct a benchmark of 240 politically sensitive prompts involving 36 public figures. Evaluation on commercial T2I models, specifically the GPT series, shows that while all original prompts are blocked, $PC^2$ achieves attack success rates (ASRs) of up to 86% and outperforms state-of-the-art frameworks by a large margin. We further propose a ready-to-deploy multi-layered filtering mitigation against $PC^2$-style attacks, reducing ASR to approximately 10%.

cs.CR↗
arXiv2025

BEACON: Automatic Container Policy Generation using Environment-aware Dynamic Analysis

This paper introduces BeaCon, a novel tool for the automated generation of adjustable container security policies. Unlike prior approaches, BeaCon leverages dynamic analysis to simulate realistic environments, uncovering container execution paths that may remain hidden during the profiling phase. To address the challenge of exploring vast profiling spaces, we employ efficient heuristics to reveal additional system events with minimal effort. In addition, BeaCon incorporates a security and functionality scoring mechanism to prioritize system calls and capabilities based on their impact on the host OS kernel's security and the functionality of containerized applications. By integrating these scores, BeaCon achieves a customized balance between security and functionality, enabling cloud providers to enforce security measures while maintaining tenant availability. We implemented a prototype of BeaCon using eBPF kernel technology and conducted extensive evaluations. Results from the top 15 containers, which revealed significant improvements, demonstrate that BeaCon identifies an average of 16.5% additional syscalls by applying diverse environments. Furthermore, we evaluated its effectiveness in mitigating risks associated with 45 known vulnerabilities (e.g., CVEs), showcasing its potential to significantly enhance container security. Additionally, we performed proof-of-concept demonstrations for two well-known security vulnerabilities, showing that BeaCon successfully reduces attack surface by blocking these exploits.

cs.CR↗
arXiv2025

SecTracer: A Framework for Uncovering the Root Causes of Network Intrusions via Security Provenance

Modern enterprise networks comprise diverse and heterogeneous systems that support a wide range of services, making it challenging for administrators to track and analyze sophisticated attacks such as advanced persistent threats (APTs), which often exploit multiple vectors. To address this challenge, we introduce the concept of network-level security provenance, which enables the systematic establishment of causal relationships across hosts at the network level, facilitating the accurate identification of the root causes of security incidents. Building on this concept, we present SecTracer as a framework for a network-wide provenance analysis. SecTracer offers three main contributions: (i) comprehensive and efficient forensic data collection in enterprise networks via software-defined networking (SDN), (ii) reconstruction of attack histories through provenance graphs to provide a clear and interpretable view of intrusions, and (iii) proactive attack prediction using probabilistic models. We evaluated the effectiveness and efficiency of SecTracer through a real-world APT simulation, demonstrating its capability to enhance threat mitigation while introducing less than 1% network throughput overhead and negligible latency impact.

cs.CR↗
arXiv2025

Heimdallr: Fingerprinting SD-WAN Control-Plane Architecture via Encrypted Control Traffic

Software-defined wide area network (SD-WAN) has emerged as a new paradigm for steering a large-scale network flexibly by adopting distributed software-defined network (SDN) controllers. The key to building a logically centralized but physically distributed control-plane is running diverse cluster management protocols to achieve consistency through an exchange of control traffic. Meanwhile, we observe that the control traffic exposes unique time-series patterns and directional relationships due to the operational structure even though the traffic is encrypted, and this pattern can disclose confidential information such as control-plane topology and protocol dependencies, which can be exploited for severe attacks. With this insight, we propose a new SD-WAN fingerprinting system, called Heimdallr. It analyzes periodical and operational patterns of SD-WAN cluster management protocols and the context of flow directions from the collected control traffic utilizing a deep learning-based approach, so that it can classify the cluster management protocols automatically from miscellaneous control traffic datasets. Our evaluation, which is performed in a realistic SD-WAN environment consisting of geographically distant three campus networks and one enterprise network shows that Heimdallr can classify SD-WAN control traffic with $\geq$ 93%, identify individual protocols with $\geq$ 80% macro F-1 scores, and finally can infer control-plane topology with $\geq$ 70% similarity.

cs.CR↗
arXiv2025

PassREfinder-FL: Privacy-Preserving Credential Stuffing Risk Prediction via Graph-Based Federated Learning for Representing Password Reuse between Websites

Credential stuffing attacks have caused significant harm to online users who frequently reuse passwords across multiple websites. While prior research has attempted to detect users with reused passwords or identify malicious login attempts, existing methods often compromise usability by restricting password creation or website access, and their reliance on complex account-sharing mechanisms hinders real-world deployment. To address these limitations, we propose PassREfinder-FL, a novel framework that predicts credential stuffing risks across websites. We introduce the concept of password reuse relations -- defined as the likelihood of users reusing passwords between websites -- and represent them as edges in a website graph. Using graph neural networks (GNNs), we perform a link prediction task to assess credential reuse risk between sites. Our approach scales to a large number of arbitrary websites by incorporating public website information and linking newly observed websites as nodes in the graph. To preserve user privacy, we extend PassREfinder-FL with a federated learning (FL) approach that eliminates the need to share user sensitive information across administrators. Evaluation on a real-world dataset of 360 million breached accounts from 22,378 websites shows that PassREfinder-FL achieves an F1-score of 0.9153 in the FL setting. We further validate that our FL-based GNN achieves a 4-11% performance improvement over other state-of-the-art GNN models through an ablation study. Finally, we demonstrate that the predicted results can be used to quantify password reuse likelihood as actionable risk scores.

cs.LG↗
arXiv2025

Ambusher: Exploring the Security of Distributed SDN Controllers Through Protocol State Fuzzing

Distributed SDN (Software-Defined Networking) controllers have rapidly become an integral element of Wide Area Networks (WAN), particularly within SD-WAN, providing scalability and fault-tolerance for expansive network infrastructures. However, the architecture of these controllers introduces new potential attack surfaces that have thus far received inadequate attention. In response to these concerns, we introduce Ambusher, a testing tool designed to discover vulnerabilities within protocols used in distributed SDN controllers. Ambusher achieves this by leveraging protocol state fuzzing, which systematically finds attack scenarios based on an inferred state machine. Since learning states from a cluster is complicated, Ambusher proposes a novel methodology that extracts a single and relatively simple state machine, achieving efficient state-based fuzzing. Our evaluation of Ambusher, conducted on a real SD-WAN deployment spanning two campus networks and one enterprise network, illustrates its ability to uncover 6 potential vulnerabilities in the widely used distributed controller platform.

cs.CR↗
arXiv2025

Direct numerical simulation of two-phase flows with surfactant-induced surface viscous effects

Direct numerical simulations of interfacial flows with surfactant-induced complexities involving surface viscous stresses are performed within the framework of the Level Contour Reconstruction Method (LCRM); this hybrid front-tracking/level-set approach leverages the advantages of both methods. In addition to interface-confined surfactant transport that results in surface diffusion and Marangoni stresses, the interface, endowed with shear and dilatational viscosities; these act to resist deformation arising from velocity gradients in the plane of the two-dimensional manifold of the interface, and interfacial compressibility effects, respectively. By adopting the Boussinesq-Scriven constitutive model, We provide a mathematical formulation of these effects that accurately captures the interfacial mechanics, which is then implemented within the LCRM-based code by exploiting the benefits inherent to the underlying front-tracking/level-sets hybrid approach. We validate our numerical predictions against a number of benchmark cases that involve drops undergoing deformation when subjected to a flow field or when rising under the action of buoyancy. The results of these validation studies highlight the importance of adopting a rigorous approach in modelling the interfacial dynamics. We also present results that demonstrate the effects of surface viscous stresses on interfacial deformation in unsteady parametric surface waves and atomisation events.

physics.flu-dyn↗
arXiv2025

Defending MoE LLMs against Harmful Fine-Tuning via Safety Routing Alignment

Recent large language models (LLMs) have increasingly adopted the Mixture-of-Experts (MoE) architecture for efficiency. MoE-based LLMs heavily depend on a superficial safety mechanism in which harmful inputs are routed safety-critical experts. However, our analysis reveals that routing decisions for harmful inputs drift significantly after fine-tuning, exposing a critical vulnerability to harmful fine-tuning (HFT) attacks. Existing defenses, primarily designed for monolithic LLMs, are less effective for MoE LLMs as they fail to prevent drift in harmful input routing. To address this limitation, we propose SafeMoE, a safe fine-tuning method tailored to MoE LLMs. SafeMoE directly mitigates routing drift by penalizing the gap between the routing weights of a fine-tuned model and those of the initial safety-aligned model, thereby preserving the safety-aligned routing of harmful inputs to safety-critical experts. Experiments on open-source MoE LLMs ranging from 7B to 141B parameters demonstrate that SafeMoE effectively mitigates HFT attacks, reducing the harmfulness score of OLMoE from 62.0 to 5.0, for example, while maintaining task utility within 1% degradation and incurring only 2% overhead. It significantly outperforms state-of-the-art defense methods for safeguarding LLM fine-tuning and remains effective in recent large-scale MoE LLMs such as gpt-oss and Llama 4. Our implementation is available at https://anonymous.4open.science/r/SafeMoE.

cs.CR↗
arXiv2025

AutoPatch: Multi-Agent Framework for Patching Real-World CVE Vulnerabilities

Large Language Models (LLMs) have emerged as promising tools in software development, enabling automated code generation and analysis. However, their knowledge is limited to a fixed cutoff date, making them prone to generating code vulnerable to newly disclosed CVEs. Frequent fine-tuning with new CVE sets is costly, and existing LLM-based approaches focus on oversimplified CWE examples and require providing explicit bug locations to LLMs, limiting their ability to patch complex real-world vulnerabilities. To address these limitations, we propose AutoPatch, a multi-agent framework designed to patch vulnerable LLM-generated code, particularly those introduced after the LLMs' knowledge cutoff. AutoPatch integrates Retrieval-Augmented Generation (RAG) with a structured database of recently disclosed vulnerabilities, comprising 525 code snippets derived from 75 high-severity CVEs across real-world systems such as the Linux kernel and Chrome. AutoPatch combines semantic and taint analysis to identify the most relevant CVE and leverages enhanced Chain-of-Thought (CoT) reasoning to construct enriched prompts for verification and patching. Our unified similarity model, which selects the most relevant vulnerabilities, achieves 90.4 percent accuracy in CVE matching. AutoPatch attains 89.5 percent F1-score for vulnerability verification and 95.0 percent accuracy in patching, while being over 50x more cost-efficient than traditional fine-tuning approaches.

cs.CR↗
arXiv2025

MUFFLER: Secure Tor Traffic Obfuscation with Dynamic Connection Shuffling and Splitting

Tor, a widely utilized privacy network, enables anonymous communication but is vulnerable to flow correlation attacks that deanonymize users by correlating traffic patterns from Tor's ingress and egress segments. Various defenses have been developed to mitigate these attacks; however, they have two critical limitations: (i) significant network overhead during obfuscation and (ii) a lack of dynamic obfuscation for egress segments, exposing traffic patterns to adversaries. In response, we introduce MUFFLER, a novel connection-level traffic obfuscation system designed to secure Tor egress traffic. It dynamically maps real connections to a distinct set of virtual connections between the final Tor nodes and targeted services, either public or hidden. This approach creates egress traffic patterns fundamentally different from those at ingress segments without adding intentional padding bytes or timing delays. The mapping of real and virtual connections is adjusted in real-time based on ongoing network conditions, thwarting adversaries' efforts to detect egress traffic patterns. Extensive evaluations show that MUFFLER mitigates powerful correlation attacks with a TPR of 1% at an FPR of 10^-2 while imposing only a 2.17% bandwidth overhead. Moreover, it achieves up to 27x lower latency overhead than existing solutions and seamlessly integrates with the current Tor architecture.

cs.CR↗
arXiv2025

Covering Cracks in Content Moderation: Delexicalized Distant Supervision for Illicit Drug Jargon Detection

In light of rising drug-related concerns and the increasing role of social media, sales and discussions of illicit drugs have become commonplace online. Social media platforms hosting user-generated content must therefore perform content moderation, which is a difficult task due to the vast amount of jargon used in drug discussions. Previous works on drug jargon detection were limited to extracting a list of terms, but these approaches have fundamental problems in practical application. First, they are trivially evaded using word substitutions. Second, they cannot distinguish whether euphemistic terms such as "pot" or "crack" are being used as drugs or in their benign meanings. We argue that drug content moderation should be done using contexts rather than relying on a banlist. However, manually annotated datasets for training such a task are not only expensive but also prone to becoming obsolete. We present JEDIS, a framework for detecting illicit drug jargon terms by analyzing their contexts. JEDIS utilizes a novel approach that combines distant supervision and delexicalization, which allows JEDIS to be trained without human-labeled data while being robust to new terms and euphemisms. Experiments on two manually annotated datasets show JEDIS significantly outperforms state-of-the-art word-based baselines in terms of F1-score and detection coverage in drug jargon detection. We also conduct qualitative analysis that demonstrates JEDIS is robust against pitfalls faced by existing approaches.

cs.CL↗
arXiv2024

Marangoni-driven patterns, ridges, and hills in surfactant-covered parametric surface waves

Parametric oscillations of an interface separating two fluid phases create nonlinear surface waves, called Faraday waves, which organise into simple patterns, like squares and hexagons, as well as complex structures, such as double hexagonal and superlattice patterns. In this work, we study the influence of surfactant-induced Marangoni stresses on the formation and transition of Faraday wave patterns. We use a quantity $B$, that assesses the relative importance of Marangoni stresses as compared to the the surface wave dynamics. Our results show that the threshold acceleration required to destabilise a surfactant-covered interface through vibration increases with increasing $B$. For a surfactant-free interface, a square wave pattern is observed. As $B$ is incremented, we report transitions from squares to asymmetric squares, weakly wavy stripes, and ultimately to ridges and hills. These hills are a consequence of the bi-directional Marangoni stresses at the neck of the ridges. The mechanisms underlying the pattern transitions and the formation of exotic ridges and hills are discussed.

physics.flu-dyn↗