Results for “Again”

arXiv (research papers) Hao Yang SI

A Cross-Modal Prompt Injection Attack against Large Vision-Language Models with Image-Only Perturbation

Large vision-language models (LVLMs) have emerged as a powerful paradigm for multimodal intelligence, but their growing deployment also expands the attack surface of prompt injection. Despite this growing concern, existing attacks still suffer from a critical limitation: the injected prompt for one modality only steers the model's interpretation of that singular input. Alternatively, these attacks remain multimodal but fail to achieve cross-modal prompt perturbation. To…

Read at arXiv (research papers)

arXiv (research papers) Tri Cao SI

WARD: Adversarially Robust Defense of Web Agents Against Prompt Injections

Web agents can autonomously complete online tasks by interacting with websites, but their exposure to open web environments makes them vulnerable to prompt injection attacks embedded in HTML content or visual interfaces. Existing guard models still suffer from limited generalization to unseen domains and attack patterns, high false positive rates on benign content, reduced deployment efficiency due to added latency at each step, and vulnerability to adversarial attacks that…

Read at arXiv (research papers)

arXiv (research papers) George Koomullil SI

Proof-Carrying Certificates for LLM Pipelines: A Trust-Boundary Architecture

We present a framework for verifying the deterministic structured computations surrounding a large language model rather than the model itself, extending a Lean 4 trust-boundary architecture to the generic interfaces of modern LLM pipelines. Certificate validity is a Lean 4 kernel type-check plus a sorry-free transitive axiom audit against the trusted set {propext, Classical.choice, Quot.sound}; other assumptions are declared and partitioned by tier (mathematical…

Read at arXiv (research papers)

Journal articles (OpenAlex) Kawsar Uddin Mahmud · Global Responsibility to Protect Geopolitics

The Influence of Hegemonic Powers on the R2P: Selective Application, Norm Contestation, and the Crisis of the Liberal International Order

Abstract Responsibility to Protect (R2P) was unanimously endorsed at the 2005 World Summit and enshrined in Paragraphs 138 and 139 of the United Nations General Assembly Outcome Document. It is deemed the international community’s most ambitious normative framework for preventing and responding to genocide, war crimes, crimes against humanity, and ethnic cleansing. Yet two decades after its adoption, R2P stands profoundly delegitimised. This article argues that hegemonic…

Read at Journal articles (OpenAlex)

arXiv (research papers) Chia-Pei Janet SI

IPI-proxy: An Intercepting Proxy for Red-Teaming Web-Browsing AI Agents Against Indirect Prompt Injection

Web-browsing AI agents are increasingly deployed in enterprise settings under strict whitelists of approved domains, yet adversaries can still influence them by embedding hidden instructions in the HTML pages those domains serve. Existing red-teaming resources fall short of this scenario: prompt-injection benchmarks ship pre-built adversarial pages that whitelisted agents cannot reach, and generic LLM scanners probe the model API rather than its retrieved content. We present…

Read at arXiv (research papers)

Journal articles (OpenAlex) Daniele Conversi, Haro L Karkour, Joan Enguer · Ethnic and Racial Studies Geopolitics

Fossil fascism: the supreme form of climate obstruction?

The last decade witnessed a proliferation of literature on far-right nationalism and climate obstruction. Scholars referred to “fossil fascism” to discuss these trends. This article emphasises fossil fascism’s weaponisation of intermediary institutions to advance a gendered and racialised vision of society that feeds off fossil fuels. This aspect, we argue, distinguishes fossil fascism from other conceptualisations of the so-called “backlash” against climate policies. We…

Read at Journal articles (OpenAlex)

Journal of Democracy Jacob Mchangama, Jeff Kosseff Geopolitics

How Hate-Speech Laws Crush Dissent Everywhere

When democracies pass laws against hate speech and extremism, they are giving autocrats the cover they need for their own crackdowns. We shouldn’t let democratic norms become a blueprint for repression. Source

Read at Journal of Democracy

arXiv (research papers) Ben Kereopa-Yorke SI

Oracle Poisoning: Corrupting Knowledge Graphs to Weaponise AI Agent Reasoning

We define Oracle Poisoning, an attack class in which an adversary corrupts a structured knowledge graph that AI agents query at runtime via tool-use protocols, causing incorrect conclusions through correct reasoning. Unlike prompt injection, Oracle Poisoning manipulates the data agents reason over, not their instructions. We demonstrate six attack scenarios against a production 42-million-node code knowledge graph, providing the first empirical demonstration of knowledge…

Read at arXiv (research papers)

arXiv (research papers) Yassin H. Rassul SI

AgentShield: Deception-based Compromise Detection for Tool-using LLM Agents

Defenses against indirect prompt injection (IPI) in tool-using LLM agents share two structural weaknesses. First, they all attempt to prevent attacks rather than detect the compromises that slip through. Second, they have only been evaluated in English, leaving users of low-resource languages such as Kurdish and Arabic without tested protection. This paper addresses both gaps with AgentShield, a deception-based detection framework that places three layers of traps inside the…

Read at arXiv (research papers)

Journal articles (OpenAlex) Don McLain Gill · The Pacific Review South China Sea

Demystifying the transparency initiative in the West Philippine Sea amid China’s influence operations against the Marcos Jr. Administration

Since 2023, the transparency initiative in the West Philippine Sea has been gaining significant attention and recognition under the administration of President Ferdinand Marcos Jr. The operationalization of the transparency initiative was catalyzed by China’s illegal actions and expansionist designs for the West Philippine Sea. While much of the focus has been on China’s kinetic-oriented maneuvers at sea, this article intends to broaden the scope of analysis towards…

Read at Journal articles (OpenAlex)

arXiv (research papers) Zhichao Liu SI

WebTrap: Stealthy Mid-Task Hijacking of Browser Agents During Navigation

Browser agents are increasingly deployed in long-horizon tasks, which require executing extended action chains to accomplish user goals. However, this prolonged execution process provides attackers with more opportunities to inject malicious instructions. Existing prompt injection attacks against browser agents expose two key gaps: (1) low effectiveness, as attacks optimized for toy baselines fail to achieve end-to-end goals in real-world scenarios with complex environments…

Read at arXiv (research papers)

arXiv (research papers) Sohom Datta SI

WAAA! Web Adversaries Against Web Agents

Large language models (LLMs) are increasingly being integrated into web browsers to create agentic browsing systems that execute actions on behalf of the user. Prior work considering the security of agentic browsers focuses on indirect prompt-injection attacks. However, by failing to consider traditional web attacks, previous agentic browser threat models have a blind spot to web social engineering attacks originally designed to trick humans. In this paper, we propose the…

Read at arXiv (research papers)

Journal articles (OpenAlex) Jae‐Eun Noh · VOLUNTAS International Journal of Voluntary and Nonprofit Organizations Geopolitics

Faith and Activism against Gender-Based Violence: The Case of the Japanese Military Sexual Slavery

Abstract Faith groups played an important role in initiating and sustaining the movement against Japanese military sexual slavery in Korea. However, individual activists’ experiences related to the connection between their faith and activism have not been extensively discussed, particularly in a transnational context. Given the transnational nature of the issue and the movement, this article explores the role of faith in meaning-making practices through activism, drawing on…

Read at Journal articles (OpenAlex)

InSight Crime Sean Doherty

Mapping Out US Alleged Drug Boat Strikes

Mapping Out US Alleged Drug Boat Strikes Since they began in September 2025, the US strikes against alleged drug boats in the Caribbean and Pacific have been marked by their opacity.

Read at InSight Crime

OnCuba News OnCuba Staff

U.S. oil siege of Cuba weighs down private sector Washington seeks to aid

The U.S. oil blockade against Cuba is heavily burdening the island’s nascent private sector, where hundreds of small businesses have closed or suspended operations, while others are reinventing themselves and fighting to survive. U.S. pressure, combined with the structural crisis the country was already enduring, is creating an environment for private Cuban entrepreneurs that is as adverse as

Read at OnCuba News

arXiv (research papers) Shihao Weng SI

ARGUS: Defending LLM Agents Against Context-Aware Prompt Injection

Large Language Model (LLM) agents are increasingly deployed as task-oriented software systems that use runtime context to decide and act on behalf of users. This delegation model makes prompt injection especially dangerous: an attacker can hide a context-aware instruction inside evidence the agent must use to decide what to do. Existing benchmarks and defenses largely miss this setting. Benchmarks often use context-insensitive tasks where the user prompt already specifies…

Read at arXiv (research papers)

Journal of Democracy Khedroob Thondup Geopolitics

China’s Campaign Against Tibetan Children

Beijing is engaged in a deliberate and devastating war on Tibetan heritage, culture, and memory. And in this war, the Chinese Communist Party’s chief target is Tibetan children. Source

Read at Journal of Democracy

Phenomenal World Daniela Gabor Geopolitics

Against Carbon Shock Therapy

Two months into the US-Israeli war on Iran, fifty-nine countries attended the first high-level conference on transitioning away from fossil fuels. The attendees who gathered in Santa Marta, Colombia, agreed that decarbonization amounted to more than energy sovereignty. It meant “building a new economy . . . that empowers communities,” to cite Stientje van Veldhoven,

Read at Phenomenal World

Journal articles (OpenAlex) Jordan W. Moon, Michael Barlev, Steven L. Neuberg · American Psychologist Geopolitics

Justifying antisemitism: Political liberalism and perceptions of prejudices.

Antisemitism has surged in the United States since the onset of the Israel-Hamas War. Yet, responses from liberals-who generally vigorously defend marginalized groups against prejudice and discrimination-have been tepid, leading some to suggest that liberals tolerate antisemitism. In three preregistered experiments (N = 979), we investigate how Americans-liberals and conservatives-perceive antisemitism and whether their perceptions depend on how the antisemitism is…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) R Tal-Singer, F Borges, F Burgos et al. · American Journal of Respiratory and Critical Care Medicine Geopolitics

C37-06 Vaccines Perceptions, Capacities, and Barriers Faced in Latin American Communities: A Study From Patient Organizations and Social Community Leaders

Abstract Introduction Vaccination against preventable diseases is a cornerstone public health intervention, yet access inequalities persist across Latin America. This study examines the role of patient advocacy non-governmental organizations (NGOs) in promoting vaccination to generate evidence that informs more inclusive, culturally appropriate policies. Methodology A descriptive-exploratory cross-sectional study was conducted in October 22-December 7, 2024, via a…

Read at Journal articles (OpenAlex)

InSight Crime Victoria Dittmar Geopolitics

The InSight Take: Rubén Rocha Moya’s Alleged Sinaloa Cartel Ties Highlight Criminal Governance in Mexico

The InSight Take: Rubén Rocha Moya’s Alleged Sinaloa Cartel Ties Highlight Criminal Governance in Mexico US prosecutors have outlined a staggering system of criminal governance in an explosive indictment against the sitting governor of Mexico's Sinaloa state for alleged links to the Chapitos faction of the Sinaloa Cartel.

Read at InSight Crime

Global Voices Daria Dergacheva

The human cost of the data center push in Latin America

In this cross-border article, Global Voices is sharing stories from several countries in Latin America where communities are pushing back against AI-driven data center development.

Read at Global Voices

arXiv (research papers) Yuan Xin SI

SafeReview: Defending LLM-based Review Systems Against Adversarial Hidden Prompts

As Large Language Models (LLMs) are increasingly integrated into academic peer review, their vulnerability to adversarial hidden prompts, i.e., adversarial instructions embedded in submissions to manipulate outcomes, poses a critical threat to scholarly integrity. We propose SafeReview, a co-evolutionary adversarial training framework for defending LLM-based peer review systems against such attacks. SafeReview jointly trains a Generator model to create sophisticated attack…

Read at arXiv (research papers)

Journal articles (OpenAlex) Lorely A Chavez, Steffanie A. Strathdee · AIDS Geopolitics

From systems that harm to systems that heal: ending HIV requires confronting gender-based violence across the life course

Violence against women and girls includes physical, sexual, psychological, or verbal abuse, such as childhood sexual abuse (CSA), intimate partner violence (IPV), and reproductive coercion (RC). CSA and IPV are well established risk factors for HIV acquisition and adverse outcomes [1,2], while RC is a newer concept that refers to behaviors that intentionally interfere with an individual's reproductive autonomy (e.g., sabotaging birth control, pressuring someone to become or…

Read at Journal articles (OpenAlex)

arXiv (research papers) Kemal Bicakci SI

Making AI-Assisted Grant Evaluation Auditable without Exposing the Model

Public agencies are beginning to consider large language models (LLMs) as decision-support tools for grant evaluation. This creates a practical governance problem: the model and scoring rubric should not be exposed in a way that allows applicants to optimize against them, yet the evaluation process must remain auditable, contestable, and accountable. We propose a TEE-based architecture that helps reconcile these requirements through remote attestation. The architecture…

Read at arXiv (research papers)

arXiv (research papers) Zonghao Ying SI

AgentVisor: Defending LLM Agents Against Prompt Injection via Semantic Virtualization

Large Language Model (LLM) agents are increasingly used to automate complex workflows, but integrating untrusted external data with privileged execution exposes them to severe security risks, particularly direct and indirect prompt injection. Existing defenses face significant challenges in balancing security with utility, often encountering a trade-off where rigorous protection leads to over-defense, or where subtle indirect injections bypass detection. Drawing inspiration…

Read at arXiv (research papers)

arXiv (research papers) Priyal Deep SI

Evaluation of Prompt Injection Defenses in Large Language Models

LLM-powered applications routinely embed secrets in system prompts, yet models can be tricked into revealing them. We built an adaptive attacker that evolves its strategies over hundreds of rounds and tested it against nine defense configurations across more than 20,000 attacks. Every defense that relied on the model to protect itself eventually broke. The only defense that held was output filtering, which checks the model's responses via hardcoded rules in separate…

Read at arXiv (research papers)

arXiv (research papers) Shutong Jin SI

CapSeal: Capability-Sealed Secret Mediation for Secure Agent Execution

Modern AI agents routinely depend on secrets such as API keys and SSH credentials, yet the dominant deployment model still exposes those secrets directly to the agent process through environment variables, local files, or forwarding sockets. This design fails against prompt injection, tool misuse, and model-controlled exfiltration because the agent can both use and reveal the same bearer credential. We present CapSeal, a capability-sealed secret mediation architecture that…

Read at arXiv (research papers)

Journal of Democracy Kim Lane Scheppele Geopolitics

How Viktor Orbán Loses

Opposition leader Péter Magyar overcame the rigged system that Orbán created and used it against him. The old playbook of disinformation and dirty tricks was no match for a disciplined campaign that gave Hungarians their voice. Source

Read at Journal of Democracy

Journal of Democracy Filip Milačić Geopolitics

How to Defeat the Authoritarian Message

If democracy’s defenders want to push back against authoritarianism, they must learn to tap into patriotic language and a sense of national identity. Such emotional appeals shouldn’t be left to autocrats and demagogues. Source

Read at Journal of Democracy

Older stories