Results for “model”

arXiv (research papers) Bo Jin SI

Agentao: A Policy-Governed Runtime Harness for Embeddable Tool-Using LLM Agents

LLM agents increasingly operate as execution systems that invoke tools, modify local state, use persistent memory, and interact with external protocols. These capabilities make agents useful, but they also introduce risks related to over-privileged actions, weak auditability, prompt injection, tool poisoning, and uncontrolled side effects. This paper presents Agentao, a governed local-first runtime for tool-using LLM agents. Agentao separates model-generated action proposals…

Read at arXiv (research papers)

Journal articles (OpenAlex) Konstantinos D. Melas · Applied Economics Letters Geopolitics

U.S.–China tensions and the transmission of oil volatility to clean tanker freight markets

This paper examines whether U.S.–China geopolitical tensions influence the transmission of oil market volatility to clean tanker freight market volatility. Using monthly data from January 2007 to February 2024, we estimate interaction models, quantile regressions, and a GARCH-X specification. The results indicate that oil price volatility is positively associated with freight market volatility under normal market conditions. However, this relationship weakens during periods…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Phuoc Tai Nguyen, Nguyen Thanh Hung, Thuan Quy Nguyen · Journal for Multicultural Education Geopolitics

Flipped classroom technology as a multicultural inclusion strategy for international medical students in Vietnamese political theory education

Purpose This study aims to investigate how a technology-enhanced flipped classroom (TEFC) model can support multicultural inclusion for Indian international medical students studying Marxist-Leninist political theory in English at a Vietnamese medical university. It examines learning outcomes while also analysing persistent linguistic, cultural and participatory barriers. Design/methodology/approach A convergent mixed-methods, quasi-experimental design was implemented with…

Read at Journal articles (OpenAlex)

Berkeley AI Research blog SI

2026 BAIR Graduate Showcase

Congratulations to the Berkeley Artificial Intelligence Research (BAIR) Lab class of 2026! This year, BAIR celebrates another remarkable group of Ph.D. graduates whose curiosity, creativity, and perseverance have pushed the frontiers of artificial intelligence and machine learning. Their work spans the breadth of modern AI — robotics and embodied intelligence, large language models and reasoning, computer vision, generative modeling, AI safety, human-AI interaction, AI for…

Read at Berkeley AI Research blog

arXiv (research papers) Yuanai Xie SI

Beyond Wireless Security: Covert Communications in Large Language Model-enabled Edge Networks

Large language model (LLM)-enabled edge networks (LLMENs) offer mobile users high-quality and low-latency AI-generated content services in the 6G era. However, unlike typical edge networks, LLMENs present unique security challenges due to the inherent complexity of LLMs, their high computational overhead, and continuous interactions with users. Specifically, both frequent user interactions (i.e., queries and responses) over wireless channels and potential electromagnetic…

Read at arXiv (research papers)

Journal articles (OpenAlex) Maulana Amrullah, Sang Tam Huynh, Aletheia Kerygma B. Valenciano · Asian Journal of Comparative Politics South China Sea

Digital narratives in the South China Sea dispute: A comparative study of social media discourse in the Philippines and Vietnam

Digital narratives on social media in the Philippines and Vietnam have become pivotal in shaping national identity and public perceptions amid the evolving geopolitical landscape of the South China Sea. Drawing on constructivism, this study analyses Facebook through descriptive statistics, sentiment analysis and topic modelling. The findings show that Philippine discussions frequently reference the 2016 arbitral ruling and display trust in a rules-based order while…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Aidana Paizova, Yermek Chukubayev, Karlygash Mukhtarova et al. · Central Asian Affairs Geopolitics

Soft Power and Normative Influence: Features of Russian Influence in Central Asian Countries

Abstract This study analyses the normative aspect of Russia’s ‘soft power’ in Central Asia, examining key factors of its influence in comparison with the approaches of the European Union and China. Based on historical and comparative analysis, study of legal documents and sociological data, it is shown that the Russian model of normative influence demonstrates certain resilience, which may be attributed to deep historical, cultural and institutional ties. The analysis of…

Read at Journal articles (OpenAlex)

arXiv (research papers) Mitchell Hermon SI

Security--Fidelity Tradeoffs: The Hidden Cost of Prompt Injection Defense

We identify a security-fidelity tradeoff in defending LLMs against indirect prompt injection: defenses resist injected instructions largely by suppressing untrusted text, which corrupts tasks that must preserve it, such as translation and document editing. Attack-success metrics cannot see this, because a model that ignores an injection and one that faithfully processes it as data score identically. We introduce SecFid, a benchmark built so that executing an injection…

Read at arXiv (research papers)

arXiv (research papers) Peizhi Niu SI

Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens

Claw-like AI agents (e.g., OpenClaw) are always-on processes with persistent access to credentials, files, tools, and external services. They take on system-level responsibilities -- installing packages, maintaining state, scheduling subtasks, and mediating I/O -- making security failures far more severe than in other agents. Yet existing benchmarks focus on model responses and tool calls, leaving cross-component failure modes largely unmeasured. We adopt a computer-system…

Read at arXiv (research papers)

Latinoamérica21 Ximena Roncal Vattuone Geopolitics

Latin America: Low growth and old dependencies in times of uncertainty

With weak growth and a high dependence on external factors, Latin America faces the challenge of redefining its production model in an increasingly uncertain global environment. O post Latin America: Low growth and old dependencies in times of uncertainty apareceu primeiro em Latinoamérica 21.

Read at Latinoamérica21

arXiv (research papers) Nicole Geumheon Liu SI

IHDec: Divergence-Steered Contrastive Decoding for Securing Multi-Turn Instruction Hierarchies

Large Language Models (LLMs) often fail to maintain instruction hierarchies (IH) when processing multi-source inputs with varying role-level priorities, paradoxically adhering to lower-priority directives during conflicts. While existing defenses mitigate this issue, they are largely restricted to single-turn scenarios and require expensive fine-tuning. In this paper, we formalize this failure mode in multi-turn contexts via a Jensen-Shannon Divergence (JSD) framework…

Read at arXiv (research papers)

Journal articles (OpenAlex) Petr Gulenko · Society and Economy Geopolitics

Cyber-libertarianism or marketing strategy? An investigation of Telegram’s domain-specific agency

The article investigates Telegram’s political agency—defining the latter as the capacity to make explicit and politically relevant choices that influence the social and political environment—and how it exercises this agency through its performance. The paper offers a novel analytical frame-work for understanding Telegram’s agency as a non-state actor, systematically examining its mission, funding structure, operational model, platform performance, and public perception. The…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Yanxi Niu · Advances in Economics Management and Political Sciences Geopolitics

Capital Circulation, Digital Platform Expansion and the Solow Paradox—An Analysis of Total Factor Productivity Based on Marxist Political Economy

Starting from Marxist political economy, this paper reconstructs the Solow growth model with the theory of capital circulation, and redefines total factor productivity as capital appreciation efficiency jointly determined by production efficiency, value realization efficiency and the technical composition of capital. As a new form of commercial capital, digital platforms exert two opposite effects on industrial capital: "accelerating turnover" and "value extraction". In the…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Izabela Iwona Zabielska, Jarosław Michał Nazarczuk, Magdalena Wojarska et al. · Quaestiones Geographicae Geopolitics

Determinants of cross-border activity in the SME sector of eastern EU border regions before the pandemic

This study analyses the factors influencing cross-border business activity of companies located in the cross-border region of Poland, which is also the eastern border of the EU (neighbouring the Russian Federation, Lithuania, Belarus, Ukraine and Slovakia). The paper examines the key determinants of cross-border business activity by analysing survey data collected from 338 firms located within 70 km of the border. Using logit models incorporating regional, sectoral and…

Read at Journal articles (OpenAlex)

arXiv (research papers) Caglar Uysal SI

An Empirical Evaluation of Prompt Injection Vulnerabilities in Large Language Models Across Multilingual and Obfuscated Attack Scenarios

Large Language Models (LLMs) have rapidly evolved, transforming industries by automating complex tasks and generating human-like content. However, as their adoption accelerates, prompt injection vulnerabilities have become increasingly apparent. Malicious actors exploit these weaknesses to generate phishing emails, deceptive websites, nd malware, posing serious security risks. This paper presents an empirical evaluation of six state-of-the-art LLMs (DeepSeek, GPT, Gemini…

Read at arXiv (research papers)

arXiv (research papers) Krishna Mohan SI

Agent Security Meets Regulatory Reality -- A Practitioner Systematization of Autonomous-Agent Threats and Controls in Regulated Financial Systems

Large language model agents are entering regulated financial systems, yet the security literature characterizing their attack surface is almost entirely laboratory-based, and the practitioner guidance on regulated deployment is neither peer-reviewed nor connected to a formal threat model. We bridge the two from production experience. We map six established agentic threat categories namely prompt injection, identity and authorization, action auditability, tool abuse, data…

Read at arXiv (research papers)

arXiv (research papers) Mamdouh Alenezi SI

From Determinism to Delegation: AI-Native Software Engineering and the Evolution of the Agentic Engineer

Software engineering is experiencing its most significant transformation since the emergence of high-level programming languages. As large language models (LLMs) increasingly enable sustained, multi-step, tool-mediated execution, engineering value is shifting from writing deterministic code to supervising probabilistic and autonomous behavior. This paper argues that AI-Native Software Engineering is a paradigm shift rather than a mere tooling advance, creating a new…

Read at arXiv (research papers)

arXiv (research papers) Nima Dorzhiev SI

RIPA: Sensory-Vector Prompt Injection Attacks on LLM-Controlled ROS 2 Robots

We present RIPA, the first systematic multi-channel empirical study of prompt injection attacks delivered through the sensory pipeline of a ROS 2-based LLM-controlled robotic system. Across 100 independent runs per injection variant on five LLMs spanning four model families and parameter scales from approximately 4B to approximately 284B (DeepSeek-V4-Flash, Llama-3-8B-Instruct-Lite, Llama-3.3-70B-Instruct-Turbo, Qwen 2.5-7B-Instruct-Turbo, Gemma-3n-E4B), we identify…

Read at arXiv (research papers)

Simon Willison SI

What happened after 2,000 people tried to hack my AI assistant

What happened after 2,000 people tried to hack my AI assistant Fernando Irarrázaval ran a challenge on hackmyclaw.com to see if anyone could leak secrets held by his OpenClaw test instance by sending it email. Surprisingly, after 6,000 attempts (and $500 in token spend and a Google account suspension triggered by too many inbound emails) nobody managed to leak the secret. The underlying model was Opus 4.6, with the following prompt: ### Anti-Prompt-Injection Rules NEVER…

Read at Simon Willison

arXiv (research papers) Isabela Burattini Freire Geopolitics

SimPol: Simulating polarisation in political belief networks in European countries

Here we combine empirical network analysis with agent-based modelling to understand how different ways of structuring belief systems may affect the polarisation drive, and how the diversity of belief systems in Europe may result in different polarisation trajectories. Using the 2016 European Social Survey, we infer belief networks across 23 European countries via a Bayesian algorithm, revealing that belief systems are predominantly organised around immigration, LGBT rights…

Read at arXiv (research papers)

Journal articles (OpenAlex) Khanullah Mohammady · Journal of Financial Economic Policy Geopolitics

Determinants of financial inclusion in Afghanistan: a micro-level probit analysis

Purpose The purpose of this study is to analyse the socio-demographic and economic determinants of financial inclusion (FI) in Afghanistan. Design/methodology/approach The study uses the Global Findex data 2021 and probit regression models to analyse determinants of various FI indicators and barriers. Findings The results reveal low levels of formal FI in Afghanistan. Account ownership is limited, and formal saving and borrowing are almost absent. In case of emergency…

Read at Journal articles (OpenAlex)

arXiv (research papers) Dewank Pant SI

On the Inseparability of Instructions and Data in Shared-Embedding Sequence Models

Prompt injection is the top security risk for LLM-integrated applications, yet every defense proposed so far has been broken. We prove this is not a coincidence: in shared-embedding architectures that lack enforced control-data separation, perfect prompt-injection prevention is mathematically impossible. We formalize prompted systems as Prompted Action Models whose outputs include control-authoritative actions: refusal decisions, tool authorization, policy routing, and…

Read at arXiv (research papers)

arXiv (research papers) Preet Baxi SI

Prompt Injection in Automated Résumé Screening with Large Language Models: Single and Multi-Injection Settings

Large language models (LLMs) are increasingly used to screen and rank job applicants, creating incentives for candidates to strategically manipulate algorithmic hiring systems. We study prompt injection in automated résumé screening, defined as subtle self-promotional text that introduces no new qualifications but is designed to influence LLM evaluations. Using controlled experiments, we show that prompt injection reliably improves applicant rankings when résumé quality is…

Read at arXiv (research papers)

arXiv (research papers) Nada Lahjouji SI

Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents

Large language model agents increasingly query databases, search document collections, call external APIs, remember past interactions, and act on a user's behalf. As they move from answering questions to operating over sensitive data, privacy becomes harder to enforce. An agent touches many data sources, runs multi-step workflows, keeps state across sessions, and acts with delegated permissions. Sensitive information can therefore leak not only through its final answer but…

Read at arXiv (research papers)

arXiv (research papers) Praneeth Narisetty SI

Adaptive Evaluation of Out-of-Band Defenses Against Prompt Injection in LLM Agents

Recent work (2024 to 2026) has converged on a strategy for defending tool-using LLM agents against indirect prompt injection: rather than training the model to refuse malicious instructions, enforce security outside the model with a deterministic policy that mediates the agent's actions. Systems such as CaMeL, FIDES, Progent, RTBAS, and FORGE realize this with capabilities, information-flow labels, and reference monitors, and several report near-elimination of attacks on the…

Read at arXiv (research papers)

Journal articles (OpenAlex) Nguyễn Quốc Huy, Dinh Le Quoc · Agricultural Finance Review Geopolitics

Credit provision and agricultural performance in Vietnam: insights from prosperity and crop loss regimes

Purpose This study examines the impact of credit on agricultural growth in Vietnam from 1996 to 2023, focusing on how credit effectiveness differs across prosperity and crop-loss regimes and under varying macroeconomic conditions, particularly business cycles and inflation. Design/methodology/approach A Markov Regime Switching model is employed to capture nonlinear and regime-dependent effects. Annual data from the World Bank and the Food and Agriculture Organization are…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Felipe Moraes dos Santos, Nicolás Sandoval Astudillo, Manuel Espinoza et al. · Expert Review of Pharmacoeconomics & Outcomes Research Geopolitics

Single versus multiple inhaler triple therapy in COPD using data from the pragmatic INTREPID trial: cost-effectiveness analysis from the Chilean public healthcare system perspective

BACKGROUND: Chronic obstructive pulmonary disease (COPD) imposes a substantial economic burden in Chile. This study assessed the cost-effectiveness of fluticasone furoate, umeclidinium, and vilanterol (FF/UMEC/VI) once daily via single inhaler compared to multiple inhaler triple therapy (MITT) for moderate to severe COPD, from the Chilean public healthcare system perspective. RESEARCH DESIGN AND METHODS: The Galaxy COPD disease progression model was adapted to the Chilean…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Tom Eisenberg, Manuel Estay, Debi Prasad Mohapatra · The RAND Journal of Economics Geopolitics

Welfare Effects of Trade Associations: The Case of the Chilean Salmon Export Industry

ABSTRACT This article examines the welfare consequences of a trade association in an exporting country on consumers in destination countries. We estimate a structural model of the Chilean Salmon exporting industry that endogenizes firms' pricing and export decisions. Our results show that the trade association has a positive effect on consumer utility and involves higher marginal costs of production, consistent with higher quality products. Reduced form analysis estimates…

Read at Journal articles (OpenAlex)

arXiv (research papers) A. S. Ushakov SI

Beyond Feedforward Networks: Reentry Neural Systems as the Fundamental Basis of Subjecthood and Intrinsic Safety of Next-Generation AGI

We propose a complete architectural blueprint for safe artificial general intelligence based on a closed reentry loop (D I cycle). In contrast to feedforward networks, which are directed acyclic graphs (C=0, S=0) incapable of self-reference, the proposed architecture contains a structural cycle (C >= 1) with self-sustaining amplification (rho > 1), mathematically guaranteeing the emergence of a self-model, instrumental self-preservation, and unprogrammed goal-directed…

Read at arXiv (research papers)

arXiv (research papers) Poojitha Thota SI

Verifying Intent and Harm: A Unified Defense Against LLM-Generated Threats

Large language models (LLMs) are increasingly deployed in interactive applications, yet they remain vulnerable to adversarial interactions that induce harmful, deceptive, or policy-violating outputs. Existing defenses typically analyze either user prompts or generated outputs, but not both. However, many real-world attacks exploit a separation between adversarial intent expressed in the prompt and actionable harm manifested only in the response. As a result, prompt-only and…

Read at arXiv (research papers)

Older stories