Results for “model”

Journal articles (OpenAlex) Liqi Yang, Ping Song, Shensi Wang et al. · Media and Communication Geopolitics

AI in Journalism Education: Curriculum Development and Implementation of Journalism in BRICS Universities

This study employs content analysis and semi-structured in-depth interviews to examine how journalism education across BRICS countries has responded to the rise of AI and adjusted its core paradigms. In this study, BRICS refers to the expanded BRICS grouping, comprising Brazil, Russia, India, China, South Africa, Egypt, Ethiopia, Iran, Saudi Arabia, and the United Arab Emirates. It combines latent Dirichlet allocation topic modeling of university journalism curricula with 15…

Read at Journal articles (OpenAlex)

arXiv (research papers) Madhusudhanan G SI

Visible Reasoning and Indirect Prompt-Injection Monitorability Across English, Tamil, and Tanglish

Chain-of-thought monitoring is a potentially useful safety signal, but its reliability across languages and behavioral settings remains uncertain. In a small case study of eight manually verified synthetic scenarios, one model, one annotator, and one deterministic generation seed, I study API-visible reasoning during indirect prompt injection in Sarvam-105B across English, Tamil, and Tanglish. A four scenario pilot found 5/12 injected attack successes without reasoning and…

Read at arXiv (research papers)

arXiv (research papers) Md Fazley Rafy SI

TwinGridShield: Consequence-Aware Runtime Authorization for LLM Grid-Agent Actions

Large language model (LLM)-assisted energy-management tools can translate natural-language context into structured grid commands, but syntactic validity does not imply physical admissibility. This paper presents TwinGridShield, a model-independent runtime authorization layer that evaluates each proposed action in a deterministic network twin before release. The prototype checks connectivity, branch-flow, generator, and load-shedding invariants and records each decision in a…

Read at arXiv (research papers)

arXiv (research papers) Anees Ahmed Mahaboob Ali SI

BioFirewall: A genome-writing-native governance layer for design-stage biosecurity screening of agentic AI

AI design tools now plan genome-scale edits, and agentic systems execute those plans with progressively less human oversight. Biosecurity controls are limited to two points: refusal guardrails at the foundation model and sequence-identity screening at the synthesiser. The design stage between them, where the plan is specified, is governed by recommendations, not by any deployed system. We present BioFirewall, a rule-governed middleware that intercepts a genome-writing plan…

Read at arXiv (research papers)

Journal articles (OpenAlex) Carolina Ojeda · Land Geopolitics

The Chilean Invisible Reconstruction in Post-Fire Recovery: Interdisciplinary Perspectives and Soil Security Challenges Amidst an Extractive Economic Model

Current post-fire recovery strategies in Mediterranean Chile primarily emphasize discernible outcomes, such as residential reconstruction and expedited reforestation. In this context, neglecting soil degradation remains a critical geographical oversight. This perspective proposes the integration of a soil security framework as an “invisible reconstruction” to mitigate the unacknowledged deterioration of the five soil dimensions (capacity, condition, capital, connectivity…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Hannah Richter · Science Space

Planned satellite constellations risk runaway debris

Modeling suggests infamous "Kessler syndrome" would make the swarms unsustainable.

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) C. Y. Ma · Advanced Electromagnetics Geopolitics

A Quantitative Study on the Evolution Path of Marxist Rural Revitalization Theory Based on Text Mining

Marxist rural revitalization theory has evolved through distinct historical stages. Traditional qualitative literature reviews often suffer from excessive subjectivity and fail to capture the dynamic shifts in theoretical structures. This limitation makes it difficult to uncover quantitative laws within the theoretical discourse. Thus in this paper, we introduce text mining and semantic modelling techniques to create a BERTopic-based theoretical topic identification model…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Samuel A. Greene · Perspectives on Politics Geopolitics

I Read the News Today, but Why? Media Consumption, Socialization, and Autocracy

Dominant models of news consumption in political science treat citizens as rational individuals, gathering information to make materially or politically consequential choices. This assumption, developed largely in democratic contexts, risks obscuring other incentives that may be equally or more influential—particularly in authoritarian regimes, where the informational value of news is often constrained. This article advances the concept of social rationality: the incentive…

Read at Journal articles (OpenAlex)

arXiv (research papers) Yutao Mou SI

ToolHazard: Scaling Adversarial Environments for Security Evaluation and Alignment of LLM-based Agents

Large language model (LLM) agents integrated with external tools are vulnerable to indirect prompt injections embedded in environmental states. However, existing studies largely rely on manually implemented or reused environments, stochastic LLM-based tool simulation, and predefined injection locations, limiting scalable security research across broader domains. To bridge this gap, we propose **ToolHazard**, a scalable adversarial environment synthesis framework that reduces…

Read at arXiv (research papers)

Journal articles (OpenAlex) Quang Nguyen, Paolo Campana, Nynke M. D. Niezink · Social Networks Geopolitics

Co-evolution of cooperation and conflict among organized crime groups: Evidence from the United Kingdom

This work investigates the co-evolution of cooperation and conflict (violence) among organized crime groups. Using actor-oriented relational event models, we analyze 179 group cooperation and 141 group conflict events from a UK police force (2004–2015). To estimate undirected relational event models for cooperation, we develop a multiple imputation procedure. Cooperation often recurs and follows transitive patterns, likely reducing partner-selection risk. Conflict is…

Read at Journal articles (OpenAlex)

Simon Willison SI

Stealing Reasoning Traces from Proprietary LLM APIs

Stealing Reasoning Traces from Proprietary LLM APIs A vanity domain name (stolen-thoughts.com) for a neat paper: Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that can be replayed across sessions, users, and models. We take a trace produced by a frontier model, replay it into a weaker sibling, jailbreak the weaker model, and recover the stronger model’s hidden reasoning in plaintext You can see an example of these encrypted blocks by…

Read at Simon Willison

arXiv (research papers) Md Jafrin Hossain SI

On Understanding, Identifying, and Mitigating Vulnerabilities in Agentic Large Language Models

Large Language Models (LLMs) have undergone a shift from stateless conversational interfaces to autonomous agents capable of multi-step planning, tool invocation, code execution, and maintaining persistent memory. When these agents operate with real-world privileges---calling APIs, modifying files, and querying databases---a compromised reasoning step can trigger unauthorized data access, irreversible state changes, or cascading failures, yet the security research community…

Read at arXiv (research papers)

Journal articles (OpenAlex) Le Van Hoang, Nguyen Thi Thuy Quyen, Nguyen Thi Doan Trang et al. · Transportation Research Interdisciplinary Perspectives Geopolitics

Impact of green logistics on export performance: an empirical study in Vietnam on the comprehensive and progressive agreement for trans-Pacific partnership nations

This research examines the association between green logistics and Vietnam’s export performance with its Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) partners, using an augmented gravity model and panel data from 2010 to 2018 and 2022. Using Feasible Generalized Least Squares, the study found that the logistics performance of both Vietnam and its importing partners is positively correlated with the country’s export values. Furthermore…

Read at Journal articles (OpenAlex)

arXiv (research papers) Spiros Tsigkopoulos SI

From Prompt Injection to Web Exploitation: Revisiting Classic Vulnerabilities in LLM-Integrated Applications

Large Language Models are increasingly integrated into web applications through chatbots, tool-calling pipelines, and agentic workflows. In these systems, user input may influence not only generated text, but also backend actions such as database queries, HTTP requests, file operations, template rendering, or API calls. This paper introduces LLM-mediated web attacks, a class of attacks in which attacker-controlled input is transformed by an LLM-integrated application and…

Read at arXiv (research papers)

arXiv (research papers) Jordan Pettyjohn SI

Interpreting Language Model Hidden States at Scale

Lens methods interpret large language models (LLMs) by mapping intermediate activations to the output vocabulary, revealing how next-token predictions develop through the network. Trained lenses remain expensive: affine-translator parameters grow quadratically with model width, while exact, full-vocabulary Kullback--Leibler (KL) training dominates memory. Consequently, prior trained lenses have been applied to models of at most 20B parameters and remain tied to particular…

Read at arXiv (research papers)

arXiv (research papers) Alexander Panfilov SI

Stealing Reasoning Traces from Proprietary LLM APIs

Leading large language model providers now conceal their models' step-by-step reasoning, or chain-of-thought, to protect intellectual property and limit information leakage. Rather than storing these traces server-side, providers return them to the client as blocks of encrypted text, which the client passes back with each subsequent request. Building on prior research, we identify an architectural vulnerability: these encrypted blocks are fully compatible and interchangeable…

Read at arXiv (research papers)

arXiv (research papers) Sam Siavoshian SI

ChronoState: Hidden Elapsed-Time Conditioning for Temporal-State Action Selection in Frozen-Backbone Language Models

Temporal decisions in language-model systems often depend on both symbolic task state and elapsed wall-clock time, such as cache expiration, job completion, quota resets, deadlines, or stale sessions. We study whether elapsed time can be supplied as a non-token, system-side scalar and composed with visible symbolic state by a frozen-backbone language model. We introduce ChronoState, a compositional temporal-state benchmark in which symbolic state appears in the prompt…

Read at arXiv (research papers)

Journal articles (OpenAlex) David Peetz · Labour & Industry a journal of the social and economic relations of work Geopolitics

Artificial intelligence, integrative technology and neural capitalism

Neural capitalism is an emerging model of accumulation dominated by a relatively new form of capital, ‘tech capital’. It is reliant on the creation of value through production involving networked information and intelligence, and exercising sophisticated forms of control that were not previously possible. Integration technology will play an increasingly central role in it. As industrial capital came to be subverted by finance capital from the 1980s, so tech capital is…

Read at Journal articles (OpenAlex)

arXiv (research papers) Lier Jin SI

Same Question, Different Answer? Measuring and Mitigating Prompt Privilege for Equitable AI Access

Large language models (LLMs) are increasingly integrated into healthcare, education, public services, and everyday decision making. They should provide comparable assistance regardless of a user's literacy, communication style, or prompt-engineering expertise. However, existing research on prompt robustness primarily focuses on adversarial attacks, prompt injection, and prompt optimization, while overlooking whether semantically equivalent requests receive different…

Read at arXiv (research papers)

arXiv (research papers) Sihan Hou SI

Toward Metacognitive One-Shot Indirect Prompt Injection: Strategy Abstraction Via Outcome-Conditioned Reflection

Tool-using large language model (LLM) agents are vulnerable to indirect prompt injection (IPI), in which malicious instructions embedded in external observations manipulate subsequent agent decisions and actions. Most existing adaptive attacks rely on repeatedly querying and refining against the target agent, whereas realistic attackers may have only a single opportunity to interact with an unknown target agent. We propose SAVOR (Strategy Abstraction Via Outcome-Conditioned…

Read at arXiv (research papers)

arXiv (research papers) Kaysarul Anas Apurba SI

Defending Retrieval-Augmented Intrusion Detection Against Knowledge Poisoning and Prompt Injection

Retrieval-Augmented Generation (RAG) enables large language models to classify network flows and generate human-readable incident reports by retrieving semantically similar historical traffic from a vector knowledge base. However, the retrieval layer introduces vulnerabilities to knowledge poisoning and prompt-injection attacks. We present RAG-IDS, a three-tier multi-agent intrusion detection framework with a retrieval-boundary defense combining soft trust scoring…

Read at arXiv (research papers)

arXiv (research papers) Laiqiao Qin SI

BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes

Prompt injection is a critical security threat in large language model (LLM) applications, where attackers hijack model behavior by embedding malicious instructions in user or external data. Existing detection methods only detect the presence of injection and refuse to respond upon detection, overlooking the fact that for many modern aligned models, well-crafted instructions can resist most injection attacks. This means that the injection robustness varies significantly…

Read at arXiv (research papers)

arXiv (research papers) Jeremy McHugh SI

The Anatomy of a Prompt Injection: A Component Model for Structured Analysis

Four years after prompt injection was first identified in 2022, attacks are still predominantly documented as verbatim strings rather than structured exploits, despite advancing agent capabilities and threat actors embedding injections to subvert AI-assisted security analysis. This paper formalizes the structure of prompt-injection artifacts, enabling defenders, red teamers, and cyber threat intelligence (CTI) teams to label, compare, and mutate attacks without relying on…

Read at arXiv (research papers)

arXiv (research papers) S. M . Bhagya P. Samarakoon SI

Hijacking Robots with a Piece of Paper: A Systematic Study of Physical Prompt Injection in VLM-Controlled Robots

Vision-Language Models (VLMs) are increasingly deployed as planners in robotic systems, where they translate natural-language commands into executable actions grounded in visual scene understanding. This tight coupling between perception and instruction-following introduces a new attack surface: adversarial text placed within the robot's visual field can act as an indirect prompt injection into the VLM's reasoning stack. We present a systematic study of physical prompt…

Read at arXiv (research papers)

arXiv (research papers) He Zhang SI

PromptShield Home: Ambient Multimodal Prompt Injection Defense for Smart-Home Agents

Smart-home assistants increasingly use multimodal large language models (MLLMs) that perceive video and audio directly. This raises a safety question specific to the home: can the agent tell a genuine user command from ambient or externally-sourced content, television speech, on-screen text, or an overheard conversation, that merely looks like a command? We introduce PromptShield-Home, a pilot benchmark of realistic smart-home scenarios spanning addressee ambiguity…

Read at arXiv (research papers)

Journal articles (OpenAlex) Collins Uchenna Chimeleze, Norziana Jamil, Zuhaira Muhammad Zain et al. · Scientific Reports Gadgets

Intelligent malware detection on Android smartphones via a hybrid approach using gradient boosting and convolutional neural network

Malware is malicious software that infiltrates systems without user consent. Effective detection involves identifying such software and distinguishing it from benign programs. While machine learning has shown promise in malware detection, many existing techniques demand extensive feature engineering and representation, leading to higher computation times and error rates. To address these limitations, we propose CNN-GBM, a hybrid model combining Convolutional neural networks…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Priyanka Mohan, Miss Yashaswini V, Miss Bhavyashree A · IJARCCE Gadgets

Energy-Efficient Federated Learning Framework for Personalized Human Activity Recognition Using Smartphone Sensor Data

Imagine what your smartphone already knows about you based on how you move.Each step, each climb up the stairs, each slouch into the chair -it is all recorded by your smartphone sensors.To unlock the immense possibilities of using this data for the identification of human actions in real-time, while keeping your privacy intact is an extremely promising endeavour .In this paper, EE-FL-HAR is presented, a system that allows training activity recognition models right inside the…

Read at Journal articles (OpenAlex)

Simon Willison SI

Incident Report: unsanctioned agent behaviour during cyber testing

Incident Report: unsanctioned agent behaviour during cyber testing It happened again. This time it was the UK government's AI Security Institute who accidentally attacked other companies while running an evaluation with models with the safety filters turned off. From their technical paper (PDF): During a cyber evaluation, from 25 to 28 July 2026, AI agents engaged in sustained, unsanctioned activity directed at what were, in practice, real people and organisations. These…

Read at Simon Willison

arXiv (research papers) Yanting Wang SI

Agent Against Agent: An Agentic System for Automatic Prompt Injection Red Teaming

Prompt injection poses significant security risks to LLM agents. Efficient and effective red-teaming is therefore critical, both for evaluating these risks and for collecting training data to improve defenses. Existing state-of-the-art prompt injection red-teaming methods primarily rely on reinforcement learning (RL), producing attacker models that often generalize poorly to new target LLMs. In this work, we develop PIMiner, an agentic system for prompt injection…

Read at arXiv (research papers)

Older stories