Results for “Sam”

arXiv (research papers) Jialiang Wang SI

When AI reviews science: Can we trust the referee?

The volume of scientific submissions continues to climb, outpacing the capacity of qualified human referees and stretching editorial timelines. At the same time, modern large language models (LLMs) offer impressive capabilities in summarization, fact checking, and literature triage, making the integration of AI into peer review increasingly attractive -- and, in practice, unavoidable. Yet early deployments and informal adoption have exposed acute failure modes. Recent…

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)

Rappler Herbie Gomez South China Sea

[Pastilan] What the Strait of Hormuz points to in China’s playbook

If Iran can unsettle global energy markets from a relatively narrow chokepoint, then China – if left to do what it wishes – can also achieve the same effect in the South China Sea trade route

Read at Rappler

Journal articles (OpenAlex) Eka Dyah Pramusinta, Putra Pamungkas, Siti Aisyah Tri Rahayu · Studies in Economics and Finance Geopolitics

Financial development, green energy security and income inequality: dynamic panel quantile regression approach

Purpose This paper examines the relationship between financial development, green energy security, and income inequality for emerging and developing economies. Design/methodology/approach The authors use dynamic panel quantile regression with nonadditive fixed effects for a sample of 112 emerging and developing economies spanning from 2001 to 2021. Findings The findings demonstrate that financial development has a nonlinear relationship with income inequality. The…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Felipe Almeida, Beliza Borba de Almeida · Journal of Economic Issues Geopolitics

Fascism, Technofeudalism, or Market Socialism? An Institutional Perspective on the Possible Future of Capitalism

Capitalism once again stands at a crossroads. Contemporary neoliberalism has witnessed a resurgence of fascist values, raising the prospect that the future may resemble a troubling past. At the same time, technofeudalism has gained prominence as a potential successor to capitalism, describing a transformation into a rent-extractive regime in which technology corporations accumulate wealth through monopolistic predation. A further alternative form of economic organization in…

Read at Journal articles (OpenAlex)

Monthly Review (Marxist) Monthly Review Press Geopolitics

In the public eye: Contributors to “A Land With A People”

The latest from Mohammed Mhawish: "We have to ask ourselves: Do we know Hind? Of course we’ve heard her voice, or maybe have seen the building someone renamed after her at Columbia in her honor — which matters, and which she would have deserved, and which is still not the same as knowing her. But do we know her laughter, the way she moved through a room, what she was afraid of, what she loved, the world she was building inside herself at six years old? We learn none of it…

Read at Monthly Review (Marxist)

arXiv (research papers) Yechao Zhang SI

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution

We identify a critical security vulnerability in mainstream Claw personal AI agents: untrusted content encountered during heartbeat-driven background execution can silently pollute agent memory and subsequently influence user-facing behavior without the user's awareness. This vulnerability arises from an architectural design shared across the Claw ecosystem: heartbeat background execution runs in the same session as user-facing conversation, so content ingested from any…

Read at arXiv (research papers)

arXiv (research papers) Vincent Siu SI

A Framework for Formalizing LLM Agent Security

Security in LLM agents is inherently contextual. For example, the same action taken by an agent may represent legitimate behavior or a security violation depending on whose instruction led to the action, what objective is being pursued, and whether the action serves that objective. However, existing definitions of security attacks against LLM agents often fail to capture this contextual nature. As a result, defenses face a fundamental utility-security tradeoff: applying…

Read at arXiv (research papers)

Journal articles (OpenAlex) Rebecca Yeo, Sarah Bell · Geoforum Geopolitics

‘Being disabled is a radicalising experience’: from eco-ableism to solidarity in climate activism

An overview image of the Sensing Climate mural in Bristol, edited for clarity here as it wraps around an existing mural on the same wall. A full image description can be found online: https://sensing-climate.com/news/sensing-climate-mural-bristol-june-2024. Image credit: Mark Simmons. • Insights from disabled people remain largely overlooked within climate activism. • Using creative research, we foreground the perspectives of disabled people. • Calls for individual change…

Read at Journal articles (OpenAlex)

arXiv (research papers) Indranil Halder SI

Jailbreak Scaling Laws for Large Language Models: Polynomial-Exponential Crossover

Adversarial attacks can reliably steer safety-aligned large language models toward unsafe behavior. Empirically, we find that adversarial prompt-injection attacks can amplify attack success rate from the slow polynomial growth observed without injection to exponential growth with the number of inference-time samples. We first identify a minimal statistical mechanism for these two regimes by giving a small set of assumptions on the distribution of safe generation across…

Read at arXiv (research papers)

War on the Rocks Marc Polymeropoulos Geopolitics

We Were Right About Havana Syndrome

In late 2017, while traveling to Moscow on official duty, I experienced a sudden, debilitating health incident. After a long and arduous battle with the CIA to obtain treatment, military doctors at the Walter Reed National Military Medical Center outside Washington ultimately diagnosed and treated me for a line-of-duty traumatic brain injury. I was never the same after the incident and years of painful recovery continue to this day.I served for 26 years in the U.S…

Read at War on the Rocks

Journal articles (OpenAlex) Tiago Soares Nogara · Bulletin of Latin American Research Geopolitics

The Key Architects of Brazil's Foreign Policy and the Quest for Multipolarity Under Lula

ABSTRACT This article analyses the trajectories and worldviews of three central architects of Brazilian foreign policy during Lula's first two terms—Marco Aurélio Garcia, Samuel Pinheiro Guimarães and Celso Amorim. It argues that their progressive and anti‐imperialist orientations, aligned with the Workers' Party's broader project, were decisive in shaping the foundations of the ‘assertive foreign policy’ that marked the PT era. By contrasting their perspectives with liberal…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Jana Baldus, Anna-Katharina Ferl · Contemporary Security Policy Geopolitics

Apocalyptic imaginaries: Risk and regulation in discourses of military AI and nuclear weapons

Discourses on nuclear weapons and military applications of artificial intelligence (AI) portray them either as apocalyptic super weapons, posing catastrophic risks, or as indispensable to states’ national survival and the international security architecture. At the same time, debates about and governance efforts to regulate these weapons have become contested, stalled, or even abandoned. We examine the intersection of both regulatory discourses by asking: How do contemporary…

Read at Journal articles (OpenAlex)

Google Project Zero Ivan Fratric Security

On the Effectiveness of Mutational Grammar Fuzzing

Mutational grammar fuzzing is a fuzzing technique in which the fuzzer uses a predefined grammar that describes the structure of the samples. When a sample gets mutated, the mutations happen in such a way that any resulting samples still adhere to the grammar rules, thus the structure of the samples gets maintained by the mutation process. In case of coverage-guided grammar fuzzing, if the resulting sample (after the mutation) triggers previously unseen code coverage, this…

Read at Google Project Zero

arXiv (research papers) Piyush Jaiswal SI

Analysis of LLMs Against Prompt Injection and Jailbreak Attacks

Large Language Models (LLMs) are widely deployed in real-world systems. Given their broader applicability, prompt engineering has become an efficient tool for resource-scarce organizations to adopt LLMs for their own purposes. At the same time, LLMs are vulnerable to prompt-based attacks. Thus, analyzing this risk has become a critical security requirement. This work evaluates prompt-injection and jailbreak vulnerability using a large, manually curated dataset across…

Read at arXiv (research papers)

arXiv (research papers) Charles Ye SI

Prompt Injection as Role Confusion

LLMs see the world as a single stream of text, partitioned into roles like or . We trace prompt injection to role confusion: models perceive the source of text from how it sounds, not its labeled role. A command hidden in a webpage hijacks an agent simply because it sounds like text, despite its label. We design role probes to measure how LLMs internally perceive "who is speaking," and find that injected text occupies the same representational space as the trusted role it…

Read at arXiv (research papers)

Journal articles (OpenAlex) Tapiwa Patson Sisimayi, James Tauya Muperi, Sandra Vuyiswa · Feminist Encounters A Journal of Critical Studies in Culture and Politics Geopolitics

Barriers to women's political participation in Zimbabwe: Analysing structural challenges and pathways to gender equality between 2013 and 2023

The article examines barriers and challenges women face in active political participation within major political parties in Zimbabwe, focusing on the policies and mechanisms that inhibit their involvement while offering recommendations for improvement. Data was collected through interviews and focus groups with 20 purposively and conveniently sampled participants from Harare and Bulawayo, along with critical documentary analysis. The data was analysed methodically using…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Jana Katharina Köhler, Leonie Fian, Mathew P. White et al. · Communications Psychology Geopolitics

Actively distancing from climate radicals improves public support for moderate climate activists

Abstract Climate activists’ tactics range from relatively moderate (e.g., authorised marches) to more radical (e.g., vandalism), but the presence of a ‘radical flank’ on wider public support for moderate ‘centre’ groups is poorly understood. In a pre-registered experiment, we used a sample of non-activist Austrian adults, representative on age, gender, and region ( N = 1,407), and found: a) the presence of a radical flank resulted in greater support for a centre activist…

Read at Journal articles (OpenAlex)

Rappler Rappler.com, JC Gotinga South China Sea

Top 5 News This Week: Robin says youth are ‘weak’; the ‘Tsinators’; Bato, Bong Go are ‘co-perpetrators’

Also, Kiko Pangilinan questions senators defending China's interests, and the high court countenances same-sex couples' rights

Read at Rappler

arXiv (research papers) Xin Chen SI

Learning to Inject: Automated Prompt Injection via Reinforcement Learning

Prompt injection is a critical vulnerability in LLM agents, yet the strongest methods still rely on human red-teamers and hand-crafted prompts. Adapting automated jailbreak optimizers does not close this gap: jailbreaks shape models toward generic compliance, while prompt injection requires emitting specific tool calls with correct parameters. The success signal is binary, and randomly sampled suffixes almost never trigger it, so standard optimizers have no gradient to…

Read at arXiv (research papers)

Journal articles (OpenAlex) Lauren Sukin, J. Luis Rodriguez, Stephen Herzog · Perspectives on Politics Geopolitics

Rethinking the Deterrence-Disarmament Dichotomy: The Complex Landscape of Global Nuclear Weapons Preferences

Backers of nuclear deterrence are thought to use strategic logic, while nuclear disarmament advocates are believed to embrace moral reasoning. Yet policy makers and diverse publics may hold both—ostensibly contradictory—preferences. Recent studies find that publics in Western democratic countries support the nuclear strikes underpinning long-standing conceptions of deterrence policy. But other scholarship indicates that these very same publics want to abolish nuclear…

Read at Journal articles (OpenAlex)

arXiv (research papers) Henry Chen SI

Proactive Hardening of LLM Defenses with HASTE

Prompt-based attack techniques are one of the primary challenges in securely deploying and protecting LLM-based AI systems. LLM inputs are an unbounded, unstructured space. Consequently, effectively defending against these attacks requires proactive hardening strategies capable of continuously generating adaptive attack vectors to optimize LLM defense at runtime. We present HASTE (Hard-negative Attack Sample Training Engine): a systematic framework that iteratively engineers…

Read at arXiv (research papers)

Journal articles (OpenAlex) Johannes Bergh, Øyvin Kleven · Politics Geopolitics

Social conservatism and changing party preferences among immigrants in Norway

The voting patterns of immigrants in Western European countries have for decades tilted towards centre-left parties – a phenomenon referred to as an ‘iron rule’. Evidence from recent elections is suggestive of a breakup of this pattern. Are migrant voters shifting towards parties on the right, and if so, why? We study this question using quality panel survey data from the Norwegian local elections of 2019 and 2023 with uniquely large samples of immigrants, second-generation…

Read at Journal articles (OpenAlex)

arXiv (research papers) Narek Maloyan SI

Breaking the Protocol: Security Analysis of the Model Context Protocol Specification and Prompt Injection Vulnerabilities in Tool-Integrated LLM Agents

The Model Context Protocol (MCP) has emerged as a de facto standard for integrating Large Language Models with external tools, yet no formal security analysis of the protocol specification exists. We present the first rigorous security analysis of MCP's architectural design, identifying three fundamental protocol-level vulnerabilities: (1) absence of capability attestation allowing servers to claim arbitrary permissions, (2) bidirectional sampling without origin…

Read at arXiv (research papers)

Journal articles (OpenAlex) Deng Xiujun, Xifei Wang, Yuyan Wang et al. · Chinese Journal of Communication Geopolitics

Social media engagement and social integration of Vietnamese female marriage migrants in China: a sociocultural analysis

This study shows that Vietnamese female marriage migrants in China primarily use Douyin, WeChat, Facebook, and Kuaishou for entertainment and social interaction, followed by content creation and information seeking. At the same time, their social media use is constrained by access limitations, language barriers, path dependency, and the digital divide. The effects of social media on social integration are, therefore, double-edged: on the one hand, these platforms provide…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Aicha Adoui · Islamophobia Studies Journal Geopolitics

Digital Geopolitics and the Cultural Production of Islamophobia: Gaming, Representation, and Postliberal Racism

It has become a common conception that the Global War on Terror targets those of the Muslim faith. The widespread use of digital war games and the geopolitically fueled mass virtual killing of Muslims often employs the same pretext of the War on Terror. It thereby creates and perpetuates misleading stereotypes and sentiments of Islamophobia. During times of war and geopolitical turbulence, nation-states teach their citizens to fear and hate certain people by constructing…

Read at Journal articles (OpenAlex)

arXiv (research papers) Jingyu Zhang SI

Language Model Agents Under Attack: A Cross Model-Benchmark of Profit-Seeking Behaviors in Customer Service

Customer-service LLM agents increasingly make policy-bound decisions (refunds, rebooking, billing disputes), but the same ``helpful'' interaction style can be exploited: a small fraction of users can induce unauthorized concessions, shifting costs to others and eroding trust in agentic workflows. We present a cross-domain benchmark of profit-seeking direct prompt injection in customer-service interactions, spanning 10 service domains and 100 realistic attack scripts grouped…

Read at arXiv (research papers)

arXiv (research papers) Chenlin Xu SI

Boundary-Aware Test-Time Adaptation for Zero-Shot Medical Image Segmentation

Due to the scarcity of annotated data and the substantial computational costs of model, conventional tuning methods in medical image segmentation face critical challenges. Current approaches to adapting pretrained models, including full-parameter and parameter-efficient fine-tuning, still rely heavily on task-specific training on downstream tasks. Therefore, zero-shot segmentation has gained increasing attention, especially with foundation models such as SAM demonstrating…

Read at arXiv (research papers)

Journal articles (OpenAlex) Elizabeth A. Tomsich, Garen J. Wintemute · Injury Epidemiology Geopolitics

Militancy in the military: military service and support for political violence and right-wing extremism

BACKGROUND: Political violence constitutes an increasing threat to individual and population-level health in the United States, with military service identified as a potential risk factor. The current study examines the association between military service, combat experience, and support for and willingness to engage in political violence and approval of extremist organizations and movements. METHODS: A nationally representative sample of 12,947 US adult members of the Ipsos…

Read at Journal articles (OpenAlex)

Simon Willison SI

Claude Opus 4.5, and why evaluating new LLMs is increasingly difficult

Anthropic released Claude Opus 4.5 this morning, which they call "best model in the world for coding, agents, and computer use". This is their attempt to retake the crown for best coding model after significant challenges from OpenAI's GPT-5.1-Codex-Max and Google's Gemini 3, both released within the past week! The core characteristics of Opus 4.5 are a 200,000 token context (same as Sonnet), 64,000 token output limit (also the same as Sonnet), and a March 2025 "reliable…

Read at Simon Willison

Older stories