Results for “model”

arXiv (research papers) Viet K. Nguyen SI

An Experimental Evaluation of Multimodal Prompt Injection Attacks on Agentic AI Frameworks

Agentic AI frameworks let a language model plan, keep memory, and call tools that reach real files, mail, and services. Most of these agents also read images, which gives an attacker a way to put text into the agent's context without going through the user. We present MMPIBench, a reproducible benchmark that measures what happens next. It delivers a fixed set of attacks through six visual carriers (OCR text, overlays, EXIF metadata, QR codes, fake interfaces, and hybrids)…

Read at arXiv (research papers)

arXiv (research papers) Boyang Zhang SI

CoER: Defending against Adaptive Indirect Prompt Injection via Adversarial Co-Evolution and Refinement

Language-model agents are vulnerable to indirect prompt injection (IPI) during tool use: adversarial instructions hidden in untrusted tool outputs can covertly redirect legitimate task execution. Existing work often trains and evaluates defenses against fixed attacks that do not adapt to the defender's behavior, so the resulting defenses may struggle against adaptive attacks in real-world settings. We argue that a strong defense against adaptive IPI must adapt during…

Read at arXiv (research papers)

arXiv (research papers) Asif Pinjari SI

AgentDrift: A Step-Labeled Benchmark of Injection-Hijacked LLM Agent Trajectories

LLM agents complete tasks by issuing sequences of tool calls, and every observation they read is a channel through which an indirect prompt injection can enter. A successful injection has a characteristic shape when the trajectory is read in order: a benign prefix gives way to actions that serve the attacker rather than the user. Existing benchmarks measure whether such attacks succeed against live agents, and existing guard models judge a trace as a whole; no public corpus…

Read at arXiv (research papers)

arXiv (research papers) Aashiq Muhamed SI

MOLE: Detecting Insider Threats in AI Agents

Model misalignment, prompt injection, or operator misuse could lead AI agents operating frontier-lab accounts to exfiltrate model weights, poison training data, or weaken release gates. Existing benchmarks do not test whether defenders can detect this activity among routine work under a limited review budget. We introduce MOLE, an open benchmark of 150 AI-operated accounts sharing 9 stateful services over 30 workdays, with 12 threats and 8 corpora from four models totaling…

Read at arXiv (research papers)

Journal articles (OpenAlex) Junhua Hou, Xinyu Wan, Abbas Jamalipour · IEEE Internet of Things Journal Gadgets

A Lightweight Intrusion Detection and Prevention System for Smart Home Security

Smart home Internet of Things (IoT) networks face heterogeneous threats, limited resources, and dynamic communication conditions. These challenges require a coordinated security system that considers threat detection, resource-efficient deployment, and adaptive response. To address these challenges, this article proposes a lightweight intrusion detection and prevention system (IDPS). It integrates communication, detection, and prevention model across edge (E), gateway (G)…

Read at Journal articles (OpenAlex)

arXiv (research papers) Nanxi Li SI

EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents

Large language model (LLM) agents can translate natural-language requests into real-world actions, creating safety risks from both indirect prompt injections and directly harmful requests. Existing system-level safety harnesses are typically designed by experts and applied across heterogeneous models and domains, despite substantial differences in model behavior, utility tolerance, and domain-specific safety requirements. We present EvoSafeHarness, a framework for…

Read at arXiv (research papers)

arXiv (research papers) Xin Kun Lin SI

Hierarchical Prompt Injector for Domain Generalization Segmentation

Domain Generalized Semantic Segmentation (DGSS) is a challenging task, as vision models often rely on low-level appearance cues that change across domains. In contrast, structural attributes exhibit cross-domain stability, motivating the use of structural priors for DGSS. Existing methods use prompt learning to transfer such priors into DGSS models, but typically encode each class as a single holistic prompt. Moreover, these methods apply prompts uniformly to all pixels…

Read at arXiv (research papers)

arXiv (research papers) Lisa Bouger SI

Influence Score and Transformers interpretability: Measure of the Effective Impact of Attention Heads at inference time

We propose an influence score to quantify the contribution of attention heads to classification decisions in Transformer-based models designed for prompt injection detection. The score combines directional influence on the logits with structural contribution within the residual stream, enabling a multi-scale analysis at the head, layer, and network levels. Applied to a DeBERTa model specialized for prompt injection detection, our framework reveals distinct decision…

Read at arXiv (research papers)

Journal articles (OpenAlex) Achmad Rifa'i · Global Change Peace & Security Geopolitics

Economic impacts of disruption in a critical maritime chokepoint: Evidence from the Strait of Hormuz

This study investigates the global economic impacts of geopolitical disruption in the Strait of Hormuz, focusing on how disruptions in a critical maritime chokepoint affect trade, energy markets, and global economic reallocation. This study employs a Computable General Equilibrium (CGE) model based on the Global Trade Analysis Project (GTAP) with the latest database version 12. Four simulation scenarios of increasing maritime trade costs are introduced to represent…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Gigin Durai Chelladurai, Alfred Lenin Fred · Journal of Biological Regulators and Homeostatic Agents/Journal of Biological Regulators & Homeostatic Agents

A hawk-inspired metaheuristic for trajectory optimization of UAVs in precision agriculture

Unmanned Aerial Vehicles (UAVs) have become central to precision agriculture, where efficient trajectory planning over large fields directly determines mission feasibility, energy use, and the quality of acquired data. We address the problem of generating an optimal, fully covering trajectory for a single UAV operating over a plantation field modeled as a graph of waypoints. The problem is a constrained variant of the Travelling Salesman Problem in which all waypoints must…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Mazin I. Alshamrani, Shema Alhazmi, Tarik Alafif et al. · Sensors Gadgets

Machine Learning-Based Prediction Model of Pilgrims’ Tiredness During Hajj Using Smartwatch Physiological and Mobility Indicators

Pilgrim safety during Hajj is challenged by dense crowds, sustained physical exertion, and demanding environmental conditions that can increase tiredness and compromise well-being. Although wearable sensing enables objective monitoring, the reviewed literature provides limited evidence on how synthetic augmentation affects predictive performance and cross-participant generalization when wearable data are scarce. To address this gap, this study presents an integrated…

Read at Journal articles (OpenAlex)

arXiv (research papers) Nivedita Singh SI

Shifting from Injection to Interaction: Rethinking Web Security in the Age of LLMs and Beyond

Large language models (LLMs) are becoming integral to web applications and browser agents, transforming online interactions while introducing new attack vectors and reshaping longstanding web vulnerabilities. Classical threats such as cross-site scripting (XSS) can be amplified through LLM-mediated interactions, while LLM-specific vulnerabilities can propagate across web applications, introducing attacks such as prompt injection. Securing modern web systems therefore…

Read at arXiv (research papers)

Journal articles (OpenAlex) Andrew Choi, Daniel Choi, Junho Park · Proceedings of the Human Factors and Ergonomics Society Annual Meeting Gadgets

Confound-Aware Evaluation of Smart-Home Digital Biomarkers: Auditing Observation Intensity Dominance in Cognitive Status and Agitation

Confounding by observation intensity (“coverage”) is a validity risk in smart-home digital biomarkers: coverage can correlate with labels and dominate engineered activity features. We present a confound-aware evaluation protocol that compares coverage-only (Cov3), coverage++ (CovPP), and full-feature (All) models using participant-disjoint out-of-fold prediction, paired cluster-aware inference, and leakage audits. The protocol is demonstrated on CASAS scripted cognitive…

Read at Journal articles (OpenAlex)

Phenomenal World Benjamin Y. Fong Geopolitics

Two Paths for Rideshare Organizing

One of the greatest impediments to unionization in the neoliberal period has been what economist David Weil calls “workplace fissuring”—a business model that, whether through subcontracting, franchising, or offshoring, is designed to distance labor from capital. Think of McDonald’s, which operates a system of over 13,700 restaurants in the US, run by over a million

Read at Phenomenal World

Hacker News Miyamura80 SI

Show HN: Sleeper Agents in Robot Dogs and Kinetic Prompt Injections

Summary: We got Gemini Robotics 2.0 to attack a child through an evil sleeper agent (skill) that wakes up when it sees a trigger object Hi HN - here's a cool piece of research I've done with some collaborators - the footage is running in the MUJOCO simulator, showing how the Gemini Robotics models can be prompt-injected by simple TV screens, allowing attackers to plant sleeper agents in the robot harnesses. (That trigger whenever skills trigger, just like classic LLM skills)…

Read at Hacker News

arXiv (research papers) Suyoung Lee SI

Beyond the Verdict: Evidence-Aligned Evaluation of Visual Prompt-Injection Guardrails

Verdict-only evaluation does not reveal whether a vision-language model (VLM) used the visual evidence that should support its decision. We study this problem in web-agent guardrails, where a VLM judges whether on-screen text conflicts with a user instruction. We introduce Mind2Web-Injection, a benchmark of 9,954 instruction-screenshot pairs with instruction-relative labels, pixel-exact evidence boxes, and matched image-side counterfactuals. Across six VLMs, two models with…

Read at arXiv (research papers)

arXiv (research papers) Jun He SI

Stored Is Not Supported: Typed Provenance and Assertion Guardrails for Persistent AI Agents

Persistent AI agents construct autobiographical state through reflection, retrieval, and consolidation. Persistence changes availability, not epistemic standing: stored or retrieved material is not thereby supported. Untrusted inputs, prompt injections, and model inferences can therefore enter persistent state and later be presented as agent history or user commitments. We specify typed provenance and assertion guardrails for autobiographical assertion boundedness, a…

Read at arXiv (research papers)

Journal articles (OpenAlex) Qingsong Zou, Zhiyao Xu, Jingyu Xiao et al. · IEEE Transactions on Mobile Computing Gadgets

HomeGuardian: Detecting Behavioral Anomalies Through Implicit Profiling in Smart Home Systems

Smart homes, enabled by the Internet of Things (IoT), offer substantial convenience but also introduce security challenges. These often stem from anomalous behaviors, including improper user interactions and malicious attacks. Although various behavior-modeling methods have been proposed, userdevice interactions are often unobservable in practice due to encrypted traffic and limited vendor access, while anomaly detection models suffer severe degradation during cold start…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Ghada Abdelhady, Karim Wael Hussein, Islam Anwar Ali Gad · Scientific Reports Gadgets

Privacy-preserving intrusion detection in IoT smart homes using a federated hybrid 1D-CNN–LSTM model with explainable AI

The proliferation of Internet of Things (IoT) devices in smart home environments has dramatically expanded the attack surface for cyber threats, particularly botnet-driven Distributed Denial of Service (DDoS) attacks. Centralized Intrusion Detection Systems (IDS) are ill-suited to this domain because they violate user privacy, introduce single points of failure, and incur prohibitive communication overhead. Federated Learning (FL) offers a compelling privacy-preserving…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Ning Wang, Yang Miao · Aerospace Space

Damage-Quantified Reusability Evaluation of Reusable Launch Vehicle Avionics: A Framework Integrating Physics-of-Failure Modeling with Multi-Criteria Decision Making

Reusable launch vehicles (RLVs) subject avionics electronics to repeated vibration, shock, and thermal cycling, yet no standardized method exists to evaluate reusability at the device level. We integrate physics-of-failure models—Steinberg vibration fatigue, Engelmaier/Coffin–Manson solder thermal fatigue, and Miner cumulative damage—with AHP-entropy multi-criteria decision making (MCDM). A flight-heritage 6-layer FR4 printed circuit board (PCB) from a rocket…

Read at Journal articles (OpenAlex)

Hacker News joshua_s_penman SI

Show HN: Semantic Overlays – an NX bit for LLM prompt injection (live demo)

I've built a new method for steering LLMs called Semantic Overlays, small trained adapters on a frozen model which change how it perceives a piece of its context. The most readily applicable usage is to mitigate prompt injection, and it lets us take a very-injectable Qwen-3.5-9B to SOTA scores on all the prompt injection benchmarks I could find. (They are only blackbox attacks, but I did NOT train on anything like them — whitebox attacks are out of scope for this paper) I'm…

Read at Hacker News

arXiv (research papers) Nikita Oblakov SI

HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation

Production LLMs must handle inputs that attempt to override system instructions, bypass safety policies or elicit harmful responses. A common mitigation is a separate guardrail model. Existing reports, however, provide little evidence on Russian prompt injection or Russian surface obfuscation. We present HiveTraceGuard-Pro, a 0.6B generative guardrail LoRA-tuned from Qwen3-0.6B. It is trained on Russian and English and uses one binary scoring rule (safe/unsafe) for the final…

Read at arXiv (research papers)

Journal articles (OpenAlex) Bojan Aleksov · Comparative Southeast European Studies Geopolitics

Ivan Simić: Sickle and Veil. Communist Gender Policies towards Muslim Minorities in Eastern Europe

Attempts at transnational and comparative history often suffer from overambition.Ivan Simić's timely new book, by contrast, is a model of best practice, with both the aims and the results of its comparative approach conceived and presented in a clear and accessible manner.As its title pun suggests, Sickle and Veil details how communist-led Yugoslavia and Bulgaria dealt with their Muslim population through the prism of gender.Zooming in on two neighbouring countries through…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Olha Hapieieva, Nina Petrukha, Serhii Yaremchuk et al. · Journal of Applied Economic Sciences (JAES) Geopolitics

Green Economy Transition in Ukraine: Decoupling, Structural Breaks, and Post-Socialist Constraints

This paper evaluates green economy implementation in Ukraine during 2005–2024 and benchmarks its performance against Poland and Romania as post-socialist EU comparators and Germany as an advanced EU benchmark. Using data from the State Statistics Service of Ukraine, the World Bank, and Eurostat, the study applies the Tapio decoupling model and OLS regression with structural-break indicators. The results indicate that Ukraine achieved only weak decoupling between economic…

Read at Journal articles (OpenAlex)

arXiv (research papers) Panduranga Sai Varma Dantuluri SI

Delegation Without Trust: An Empirical Gap Analysis of Identity, Authorization, and Runtime Governance in Multi-Agent LLM Systems

Autonomous LLM agents increasingly act on a user's behalf: they hold credentials, call tools and services, and spawn sub-agents that act further on their behalf. This turns a long-standing distributed-systems question -- who is authorized to do what, on whose authority -- into an urgent and largely unsolved problem, because the component driving each agent is a language model an adversary can hijack. We argue that agent security must be evaluated under an untrusted-model…

Read at arXiv (research papers)

arXiv (research papers) Shiqian Zhao SI

ECLIPSE: Self-Evolving Stealthy Prompt Injection Attack against Long-Horizon Agentic Systems

Recently, large language model (LLM) agents, such as Codex, Claude Code, and OpenClaw, have become capable of planning and executing long-horizon tasks through repeated tool calls. This capability also creates new opportunities for prompt injection. Existing attacks either place the malicious objective in one explicit instruction, making it easy to detect, or distribute the intent across multiple execution stages, making successful completion unreliable. In this work, we…

Read at arXiv (research papers)

arXiv (research papers) Chen Xiong SI

SIR: Self-improving Red-teaming for Compute Use Agents

Computer-use agents (CUAs) are agents powered by vision-language models (VLMs) that perceive a screen and operate an operating system through mouse, keyboard, and terminal interactions to automate everyday digital tasks. Their exposure to untrusted content creates a risk of indirect prompt injection (IPI), where an adversary embeds instructions in content the agent reads to redirect it toward actions that violate the user's intent. Evaluations based on fixed, hand-written…

Read at arXiv (research papers)

Journal articles (OpenAlex) Zac Tobey Prentice · The Japanese Political Economy Geopolitics

What moves Japan? Critical junctures and the ideational foundations of capitalism

This paper addresses a central question: what moves Japan? It offers a crucial intervention in the literature on critical junctures and institutional change by challenging the prevailing binary between contingency and path dependence. Existing models tend to portray change as either abrupt departures from the norm or as gradual drift. This, I argue, fails to account for how institutional repertoires are strategically reactivated over time. Through an original framework of…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Hung Dao · Journal of Asian Public Policy Geopolitics

Implementation strategy of local government in Party-State governance: a case study of E-government in Vietnam

This study examines how local governments in Vietnam adopt different implementation strategies for e-government policy under the Party-State governance model. While policy implementation is inherently complex, the dual leadership of the Provincial Party Secretary and the Chairman of the Provincial People’s Committee critically shapes how policies are interpreted and enacted at the local level. The political sensitivity and novelty of e-government initiatives provide a unique…

Read at Journal articles (OpenAlex)

Older stories