Results for “model”

arXiv (research papers) Yang Gao Veyon Solutions SI

How Reliable Is Your Jailbreak Judge? Calibration and Adversarial Robustness of Automated ASR Scoring

Almost every paper on LLM jailbreaks and prompt injection reports an attack-success rate (ASR), and that number is assigned not by people but by an automated judge: either a safety classifier trained for the task, or a general chat model prompted to grade. The judge is rarely checked. We check it. Using 596 human-labeled completions from the HarmBench classifier validation set, we compare the two judge families against human majority votes and then attack them. The two…

Read at arXiv (research papers)

arXiv (research papers) Thi Huyen Nguyen SI

LLM-Based Scientific Peer Review: Methods, Benchmarks, and Reliability Challenges

The rapid growth of scientific submissions has pushed traditional peer review toward its scalability limits, motivating the exploration of large language models (LLMs) as intelligent automated evaluation assistants. Although recent studies show that LLMs can generate fluent critiques and approximate reviewer scores, their reliability, robustness, and security as decision-support systems remain insufficiently understood. This survey offers a systems-level analysis of…

Read at arXiv (research papers)

Simon Willison SI

Prompt Injection as Role Confusion

Prompt Injection as Role Confusion First, I absolutely love this: This is a blog-style writeup of the paper. I wish every paper would come with one of these. Academic writing is pretty dry - the impact of a paper can be so much higher if you publish a readable version to accompany the formal one. Charles Ye, Jasmine Cui, and Dylan Hadfield-Menell present some fascinating research into the challenge of having models distinguish their own privileged text (here wrapped in role…

Read at Simon Willison

arXiv (research papers) Adam Storek SI

GIF: Locally Sound Geometric Information Flow Control for LLMs

Large language models increasingly mediate interactions between sensitive data, untrusted inputs, and privileged actions in agentic systems, creating security and privacy risks. These range from prompt injections that manipulate downstream tool use to leakage of confidential information through model outputs. Recent Information Flow Control (IFC)-based defenses show promise but lack a principled semantic foundation for reasoning about information flow through the model…

Read at arXiv (research papers)

arXiv (research papers) Yanhang Li SI

When AUC 0.998 Is Not Enough: A Candidate Evaluation Protocol for Hidden-State Probes of Indirect Prompt Injection in Multimodal Computer-Use Agents

Hidden-state probing -- a linear classifier on a frozen vision-language model's internal activations -- has emerged as an attractive evaluation tool for flagging indirect prompt injection (IPI) in multimodal computer-use agents before the agent emits a corrupted action. We argue, on a single-backbone cautionary case study (Qwen2.5-VL-7B on Mind2Web, teacher-forced replay), that a high probing AUC on a clean-vs-attack split is not, on its own, evidence of malicious-content…

Read at arXiv (research papers)

arXiv (research papers) Xingwei Zhong SI

DE-FIVE: Detecting Malicious Image Prompts via Fourier Features and Image Vector Embeddings

Vision language models (VLMs) employ both visual and textual modalities to enable advanced vision-language inference. However, incorporating visual modalities expands the attack surface of VLMs, making them more susceptible to security threats such as adversarial perturbations and indirect prompt injection, wherein crafted malicious image prompts can elicit unintended model outputs. Existing defense methods against malicious image prompts remain insufficient as they…

Read at arXiv (research papers)

arXiv (research papers) Md Anas Biswas SI

Confidently Wrong: Severity-Aware Calibration of Prompt-Injection Detectors under Attack Shift

Prompt-injection detectors are deployed as guards: a model scores an input and a downstream system trusts or blocks it on that score. I study the confidence of these scores, not only their accuracy, when the attack distribution shifts away from the clean benchmark on which the operating point was chosen. I evaluate three released detectors, ProtectAI-v2 and two Prompt-Guard-2 checkpoints, at a single source-calibrated threshold that I freeze and transport across five shifts…

Read at arXiv (research papers)

Havana Times Circles Robinson

Cuba to Move from State Enterprises to a Corporate Model

According to the Communist Party / government proposal, the state would remain a majority shareholder only in sectors considered strategic.

Read at Havana Times

Latinoamérica21 Danilo Desiderio Geopolitics

How the Global South anticipated the rules of trade in times of uncertainty

In a world shaped by fragmentation and uncertainty, Global South economies developed a degree of trade flexibility that is now becoming a model for the traditional powers. O post How the Global South anticipated the rules of trade in times of uncertainty apareceu primeiro em Latinoamérica 21.

Read at Latinoamérica21

arXiv (research papers) Zhen Zhao SI

AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents

Traditional OS security models based on "resource exposure plus permission checks" face structural challenges as LLM-driven autonomous agents acquire capabilities for planning, tool use, network access, and code execution. Once an agent runtime is compromised through prompt injection or malicious tool outputs, an attacker can compose POSIX-style resource primitives into behaviors far beyond the user's task authorization. To address this, we propose AgenticOS, an…

Read at arXiv (research papers)

arXiv (research papers) Zhengsong Zhang SI

Local LLM Agents as Vulnerable Runtimes:A Source-Code Audit of the Agent Runtime Layer

Local LLM agents such as OpenClaw and Nanobot run on end-user machines and act on host resources - the shell, filesystem, browser, stored credentials, and messaging applications - through natural-language goals. These agents have become privileged software runtimes that mediate between user intent, model outputs, and host-level actions. Existing research characterizes the landscape through prompt injection, malicious skills, marketplace risks, or black-box evaluation of…

Read at arXiv (research papers)

arXiv (research papers) Shanghao Shi SI

Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning

The integration of external tools has substantially expanded the capabilities of large language model (LLM) agents, but it also introduces new attack surfaces beyond prompt injection. In particular, cross-tool description poisoning can manipulate planner-visible tool metadata to steer an agent's trajectory, even if the poisoned tool itself is never chosen. To understand the effectiveness of existing defenses against this emerging threat, we first evaluate several…

Read at arXiv (research papers)

arXiv (research papers) Reza Soosahabi SI

Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems

Agentic AI systems increasingly rely on language-model components to interpret instructions, process external data, invoke tools, and coordinate with other agents. These capabilities make prompt-injection and jailbreak attacks more consequential, especially as attackers adopt model-guided automation to scale probing, prompt refinement, and response evaluation. This work analyzes the resulting attack-defense setting through a probabilistic model of a target system, its…

Read at arXiv (research papers)

The Guardian Communism Agence France-Press in Havana Geopolitics

Cuban economy needs ‘urgent changes’ as US blockade deepens crisis, says president

Miguel Díaz-Canel cites China and Vietnam as possible models for opening up the country’s economy Cuba’s economy needs urgent changes to overcome a crisis intensified by a US oil blockade, the president, Miguel Díaz-Canel, has said in a speech to Communist party leaders. “The situation calls for urgent and necessary changes,” Díaz-Canel told the party’s politburo in his frankest admission yet of the need to overhaul the country’s communist model.

Read at The Guardian Communism

arXiv (research papers) Gulshan Saleem SI

A Layered Security Framework Against Prompt Injection in RAG-Based Chatbots

Prompt injection is ranked as the most critical vulnerability in large language model (LLM) deployments by the OWASP Top 10 for LLM Applications, yet existing defenses operate at isolated pipeline stages and remain incomplete. Input filters cannot inspect retrieved documents, while output monitors cannot prevent malicious payloads from reaching the model. Consequently, retrieval-augmented generation (RAG) chatbots remain vulnerable to indirect injection, where a poisoned…

Read at arXiv (research papers)

arXiv (research papers) Zunchen Huang SI

Analyzing the Narration Gap in LLM-Solver Loops

Formal tools such as SAT and SMT solvers are increasingly embedded in language model reasoning pipelines when a safety or security critical question can be formulated in logic. Unlike chain of thought whose steps are sampled from the model distribution without formal guarantee, a solver produces a sound and independently verifiable answer. However, the soundness guarantee can be lost in the interaction between the solver and the model. The hybrid pipeline has three…

Read at arXiv (research papers)

arXiv (research papers) Po-Han Cheng SI

CodeSentinel: A Three-Layer Defense Against Indirect Prompt Injection in Code Contexts

Code large language models increasingly retrieve external code context from repositories, documentation, issue threads, and coding-agent environments, creating an indirect prompt-injection surface where attackers hide instructions in comments, strings, identifiers, or decoy code. We propose CodeSentinel, a three-layer inference-time sanitizer. It uses Tree-sitter to extract high-risk model-facing CST nodes, then combines syntax-guided pre-filtering, CST-guided Dynamic…

Read at arXiv (research papers)

arXiv (research papers) Yuchuan Tian SI

SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents

Tool-using language-model agents introduce security failures that go beyond unsafe text: they can disclose protected objects, write persistent memory, send messages, modify databases, or trigger harmful code and tool effects. Existing evaluations often collapse these stages into a single attack success rate, making it difficult to tell whether a model merely agreed with an attacker or actually produced observable harm. We introduce SafeClawBench, a staged benchmark for…

Read at arXiv (research papers)

arXiv (research papers) Xuelong Dai SI

MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents

Multimodal Large Language Model (MLLM)-based web agents provide practical, high-precision solutions for visual browser automation; however, they inherently expand the attack surface, introducing novel vision-based vulnerabilities. Existing adversarial evaluations targeting these agents frequently rely on permissive threat models and visually conspicuous artifacts. In this paper, we investigate a constrained vulnerability detection setting: a trusted web platform where the…

Read at arXiv (research papers)

arXiv (research papers) David Huang SI

Rapid Poison: Practical Poisoning Attacks Against the Rapid Response Framework

The Rapid Response (RR) framework, deployed in production systems, including Anthropic's ASL-3 safeguards, continuously improves jailbreak-detection classifiers. When new jailbreaks emerge that bypass these classifiers, Rapid Response generates synthetic variants for training, helping the model generalize from the new attacks and quickly adapt. We reveal that prompt injection can infiltrate this pipeline to deliver poisoned samples into the classifier's training set…

Read at arXiv (research papers)

Journal articles (OpenAlex) Fabrizio Del Carpio Delgado, Rafael Romero-Carazas, Roque Juan Espinoza Casco et al. · F1000Research Geopolitics

Temporal prediction of arrests for illicit drug trafficking and micro-distribution using ARIMA models in Peru

Introduction Illicit drug trafficking and micro-distribution represent persistent challenges affecting public security, institutional governance, and social dynamics in Peru. In this context, the aim of the study was to temporally predict trends in arrests related to illicit drug trafficking and micro-distribution using ARIMA models in Peru. Methods The research was conducted under a quantitative approach with a non-experimental, descriptive-predictive, and retrospective…

Read at Journal articles (OpenAlex)

arXiv (research papers) Aman Anifer SI

GAS-Leak-LLM: Genetic Algorithm-Based Suffix Optimization for Black-Box LLM Jailbreaking

Large Language Models (LLMs) constitute pivotal components within the AI-dominated information technology ecosystem. To mitigate risks associated with harmful or policy-violating outputs, commercial systems employ advanced alignment strategies and multi-layered content moderation mechanisms. Despite these safeguards, recent research has demonstrated that LLMs remain vulnerable to adversarial manipulation, particularly through jailbreaking and prompt injection techniques. In…

Read at arXiv (research papers)

arXiv (research papers) Zixin Rao SI

FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion

Large language model (LLM) agents increasingly rely on long-term memory to support complex task execution, user personalization, and domain adaptation. Meanwhile, emerging access-control mechanisms for LLM agents are being explored to block policy-violating requests and prevent misuse. We reveal a novel attack surface arising from agent memory operations: prohibited content that would trigger access control can be fragmented across interactions, stored in long-term memory in…

Read at arXiv (research papers)

Journal articles (OpenAlex) José Ángel Brandariz, Romina Ramos Rodríguez · International Journal for Crime Justice and Social Democracy Geopolitics

Large Fishing Dragnets Catching Small Fish: Anti-Migrant Smuggling Policies in Chile

By oversimplifying the complexity of human smuggling and facilitation phenomena, anti-smuggling policies in Global North countries have acted as large fishing dragnets that catch small fish, since they frequently focus on minor cases and punish migrants’ relatives and communities, solidarity networks and activists, as well as migrants and asylum seekers themselves. This concerning scenario shows that the anti-smuggling model that has been widespread since the early 2000s is…

Read at Journal articles (OpenAlex)

arXiv (research papers) Zhongye Liu SI

Forced Deferral: Manipulating Routing Decisions in Multimodal LLM Cascades

While multimodal large language models (MLLMs) have shown strong visual reasoning abilities, serving a large model for every query is computationally expensive. MLLM cascades mitigate this cost by first querying a weak but cheaper model and deferring to a strong model when the weak model's output is unconfident. However, since the weak model's confidence directly controls compute allocation, these systems expose a new attack surface: an adversary can manipulate confidence so…

Read at arXiv (research papers)

arXiv (research papers) Saeid Jamshidi SI

Security Engineering of OpenClaw: Analyzing Attack Surface Expansion and Trust-Boundary Violations

Agentic large language model (LLM) systems can now execute actions, not only produce text. When model outputs trigger privileged operations such as shell commands, browser automation, or external tool calls, the security problem shifts from alignment alone to system configuration and structural design. We analyze OpenClaw, a self-hosted multi-agent system in which LLM outputs can execute commands and interact with tools and services. We measure compromise probability…

Read at arXiv (research papers)

arXiv (research papers) Youngduk Kim SI

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…

Read at arXiv (research papers)

arXiv (research papers) Haowei Qian SI

Nous: An Attempt to Extract and Inject the Cognition Behind Prediction-Market Behavior

As LLM agents proliferate in prediction markets and collective decision-making, they risk a cognitive monoculture: agents built on shared foundation models produce correlated forecasts, and recent measurement finds frontier-model errors correlated at r ~ 0.77. We ask whether human cognitive diversity can be recovered from behavior and transferred to LLM agents. Nous extracts a structured eight-dimension behavioral profile from real Polymarket trading activity and injects it…

Read at arXiv (research papers)

Journal articles (OpenAlex) S Liu, Eunsoo Kim, Insoo Son · Sustainability Geopolitics

Coffee Export Competitiveness in China and Vietnam: A Comparative Gravity Analysis of Demand, Supply, and Trade Policy, 2001 to 2022

Despite geographical proximity and broadly similar agro -climatic conditions, China and Vietnam show sharply divergent coffee export performance, with Vietnam ranking as the world’s second largest exporter, while China’s exports remain modest. This study compares the determinants of their bilateral coffee exports over 2001 to 2022, using a gravity model estimated by Poisson pseudo maximum likelihood with partner and year fixed effects, a specification that retains zero trade…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) André Grimaud, Gilles Lafforgue, Luc Rougé · Economics Letters Geopolitics

Does the climate challenge justify degrowth?

Degrowth is often advocated as a response to the climate crisis, but its consistency with growth theory remains unclear. We develop an endogenous growth model of directed technical change and climate in which the economy relies on both a polluting fossil resource and a clean renewable alternative. We fully characterize the social optimum and show that accounting for climate damages may justify an initial phase of degrowth. Such a phase is more likely when fossil resources…

Read at Journal articles (OpenAlex)

Older stories