Results for “model”

Journal articles (OpenAlex) Paolo Muratore-Ginanneschi, Bayan Karimi, Jukka P. Pekola · Journal of Statistical Physics

Decay of the Survival Probability of a Local Excitation in Multi-qubit Platforms

Abstract We present a theoretical study of the survival probability of a state initially prepared in the one-particle sector of a multi-qubit system. The motivation for our work is the ongoing laboratory development of multi-qubit platforms based on superconducting circuits. Using elementary concepts of random matrix theory, we obtain analytic expressions for the survival probability in mathematical models of platforms which, albeit stylized, have been previously shown to…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Bülent Aras, Burcu Fazlioglu · German Politics Geopolitics

Germany’s Diplomatic Pulse: A Data-Driven Examination of Shifting Foreign Policy Priorities (1991–2023)

This article empirically investigates the fundamental reorientation of Germany’s foreign policy from a normative, trade-centric ‘civilian power’ to a pragmatic ‘strategic hedger’ in response to a fragmenting global order. To bridge the gap between political rhetoric and actual state behaviour, the study introduces an original large-N dataset tracking official international bilateral visits by German Chancellors from 1991 to 2023. By utilising logit regression models to…

Read at Journal articles (OpenAlex)

arXiv (research papers) Amelia Kawasaki SI

Attention-based Classification for Indirect Prompt Injection

Indirect prompt injection (IPI) attacks, in which adversarial instructions are embedded in external data processed by a language model, pose a significant threat to the integrity of deployed LLM agents. Attention Tracker (Hung et al., 2025) is a recently proposed white-box detection method that identifies prompt injection attacks by monitoring how a model's attention is distributed between user task tokens and data tokens, claiming dataset-agnostic detection via a scalar…

Read at arXiv (research papers)

arXiv (research papers) Pedro H. Barcha Correia SI

Combining Long-Term Memory and LLMs Comes at a High Cost: From Prompt Injection to Persistent Compromise

The widespread adoption of generative artificial intelligence in recent years has brought a parallel rise in prompt injection and jailbreak attacks. By embedding malicious instructions, these exploits manipulate language models to bypass intended behavior and operational boundaries. In traditional chatbots, such attacks are confined to a single session because state memory is not preserved across conversations. However, persistent memory features in modern conversational…

Read at arXiv (research papers)

arXiv (research papers) Abhishek Rai SI

Context-Aware Prompt Injection Detection via Small Language Models in RAG Pipelines: nokast-secureRAG

Local Retrieval-Augmented Generation (RAG) systems are vulnerable to indirect prompt injection and retrieval poisoning, where malicious instructions in retrieved content are mistaken for trusted input. We present nokast-secureRAG, a roughly 0.5B-parameter small language model that acts as a semantic firewall between retrieval and generation. It jointly evaluates the user query and retrieved context, classifying the pair as safe, suspicious, or malicious. The model is trained…

Read at arXiv (research papers)

Hacker News denn-gubsky SI

Decision model as prompt-injection judge: 40 of 40 on 78 texts

Article URL: https://loomcycle.dev/blog/decision-model-as-prompt-injection-judge Comments URL: https://news.ycombinator.com/item?id=50166532 Points: 1 # Comments: 0

Read at Hacker News

Hacker News Hiskias SI

Building a System One Model for Prompt Injection

Article URL: https://www.stackone.com/blog/system-one-model-prompt-injection/ Comments URL: https://news.ycombinator.com/item?id=50165645 Points: 2 # Comments: 0

Read at Hacker News

arXiv (research papers) Vasisht Duddu SI

KNOX: Hardware-assisted Regulatory Sandbox for Generative Models

Regulations require companies to run AI models in a regulatory sandbox, an isolated environment that enables monitoring model behavior by regulators, prior to deployment. But, they do not spell out the technical mechanisms to design such a sandbox. Ideally, a regulatory sandbox should show (i) compliance (checks against regulations, and deviations from intended behavior, i.e., unauthorized tool and system calls); (ii) confidentiality (protects model weights from regulators)…

Read at arXiv (research papers)

SiliconANGLE AI James Farrell SI

Trump administration orders ban on DeepSeek in federal agencies

The Department of Homeland Security has ordered artificial intelligence technology from the Chinese company DeepSeek Ltd. be removed from all federal civilian agencies. The order, published on the Federal Register website today, warned that DeepSeek’s AI models present “multiple threat vectors,” that could give the Chinese Communist Party “extensive authority” to access American government infrastructure.

Read at SiliconANGLE AI

Journal articles (OpenAlex) Richard Fu, Ajay Subramanian · Financial Management Startups

Entrepreneurial Optimism and Venture Capital: A Quantitative Analysis

ABSTRACT We quantitatively analyze the impact of entrepreneurial (EN) optimism on venture capital (VC) projects in a dynamic calibratable model. VCs and entrepreneurs (ENs) have differing beliefs and risk attitudes about projects. The equilibrium features staged VC investments, progressive vesting of the EN's stake, and intertemporal performance targets. Optimism is a major driver of VC projects. The EN's mean project quality assessment is over twice the VC's mean…

Read at Journal articles (OpenAlex)

MIT News AI Alex Ouyang | Abdul Latif Jameel Clinic for Machine Learning in Health SI

Why teaching AI about symmetry could make prenatal diagnosis more reliable

A model developed by researchers at MIT and Boston Children's Hospital adjusts scans on the fly to track fetal movement, potentially sparing families repeat hospital visits.

Read at MIT News AI

arXiv (research papers) Mengxiao Wang SI

Fake News, Real Crashes: Information Injection Attacks on LLM Financial Trading Agents

Large language model (LLM) agents now trade real capital using news as evidence, yet they cannot reliably tell fabricated headlines from real news. Because they read the same feeds under near-identical policies, a single fabricated post can push them into the same trade at the same moment. We present a systematic attack benchmark for information injection against LLM trading agents. Unlike prompt injection, which targets the model's instructions, information injection…

Read at arXiv (research papers)

arXiv (research papers) Yinxiang Wang SI

SpanGuard: Decoupling Instruction Localization from Authorization for Lightweight IPI Defense

LLM-based guardrails are effective against indirect prompt injection (IPI), but their performance depends heavily on model capability: as the guardrail model becomes smaller or weaker, detection degrades substantially. This raises a natural question: can we retain strong IPI defense while relying only on a small LLM? To understand why, we revisit what an LLM guardrail actually does. Detecting IPI requires simultaneously locating potentially injected instructions within…

Read at arXiv (research papers)

arXiv (research papers) Yuchen Chen SI

RAGLeak: Supply-Chain Privacy Backdoors in Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) enables large language models to access external knowledge during inference, but its reliance on external knowledge bases also introduces new attack surfaces. Existing data extraction attacks against RAG systems rely on either prompt injection, which heavily depends on the generator's instruction-following capability, or backdoors that reproduce or paraphrase retrieved documents but struggle to balance extraction fidelity and…

Read at arXiv (research papers)

Journal articles (OpenAlex) Shahrear Roman, Hadi Hadi, Wyclife Agumba Oluoch et al. · Proceedings of the National Academy of Sciences

Modeling the global potential of smart farming innovations

Smart farming innovations can improve agricultural productivity, resilience, and environmental sustainability, yet their real-world impact hinges on their diffusion dynamics, which in turn depend on their economic viability. We present here a global assessment of the profitability of smart farming innovation across croplands at a spatial resolution of ~5 km from the farmers' and the societal perspectives. We cover application mapping for precision agriculture, automated…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Nathaniel Furman, Albert Herrero-Parareda, Anthony Rapp et al. · Optical Materials Express

Experimental measurement of enhanced group delay in silicon photonic waveguides indicative of the frozen mode regime around the stationary inflection point [Invited]

The dispersion engineering of periodic silicon photonic waveguides presents opportunities for significant group delay enhancement compared to uniform waveguides of comparable length. We describe the spectral response characteristics for measured devices composed by three coupled waveguides and compare their properties to modeled data. These waveguides support the frozen mode regime (FMR) around near infrared wavelengths and are expected to show enhanced group delays around…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Ahmad Mohammadpour, Jalil Karimi · Acta Sociologica Geopolitics

Natiosociology: Unraveling the nationalist epistemology in Iranian sociology

This article examines the development and trajectory of Iranian sociology, highlighting its entanglement with abstraction, domination, and the erasure of difference. Imported as a discipline, Iranian sociology has historically followed a Eurocentric model of knowledge and domination, framing the realities of non-sovereign identities in Iran through a metropolitan lens, thereby marginalizing peripheral identities. The analysis identifies three interrelated forms of…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) S. Suganya, T. Manikandan · African Journal of Science Technology Innovation and Development Space

Scalable and robust deep learning models for space debris detection across orbital regimes and observation platforms

Space debris poses a significant threat to space missions, requiring accurate and real-time detection across multiple orbital regimes and observation platforms. This study aims to develop a scalable YOLOv12-PSO framework for accurate space debris detection and to investigate its effectiveness in improving real-time performance, robustness, and detection accuracy across orbital environments. A publicly available space debris image dataset containing synthetic and real images…

Read at Journal articles (OpenAlex)

arXiv (research papers) Stavros Damianakis SI

SoK: PIAtlas: Decomposing the Unbounded Prompt Injection Attack Space

Prompt injection is the dominant security risk for large-language-model (LLM) applications and agents, yet the field still lacks a coherent map of how these attacks are constructed. More than twenty public datasets describe prompt injection using incompatible vocabularies, often organized around delivery vector, attacker goal, or generation method. In this Systematization of Knowledge, we present PIAtlas, a taxonomy organized instead around how the injected instruction is…

Read at arXiv (research papers)

MIT News AI Zach Winn | MIT News SI

Giving AI a “crystal clear” view of the materials world

New research can help AI models understand crystalline materials in 3D and reason about their properties.

Read at MIT News AI

Journal articles (OpenAlex) Bhagwat Mahesh, T. S. Vishwanath · Proceedings on Engineering Sciences Security

A HYBRID ANFIS-GA FRAMEWORK WITH EXPLAINABLE AI (XAI) FOR TRANSPARENT AND ACCURATE PHISHING DETECTION

Considering the fact that phishing sites remain a prevalent and serious attack on the Internet, the sphere is in dire need of the creation of smart and dynamic defense systems.We overcome this issue by providing a new hybrid neuro-evolutionary fuzzy model.This system is a careful integration of the Adaptive Neuro-Fuzzy Inference System (ANFIS) that is augmented with a Genetic Algorithm (GA) to do a global optimization and integrated Explainable AI (XAI) to make the entire…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Arepalli Peda Gopi, K. Jairam Naik, Kanusu Srinivasa Rao et al. · The Journal of Supercomputing Security

A multi-feature dueling DDQN model for robust and interpretable malware detection

Read at Journal articles (OpenAlex)

arXiv (research papers) Songping Wang SI

MobileInject: Evaluating Reasoning-to-Action Security of MLLM-Based Mobile Agents under Cross-Modal Consistency Injection

Multimodal large language model (MLLM)-based mobile agents are rapidly evolving from conversational assistants into autonomous systems that perceive, reason over, and act on smartphone environments. This capability also exposes a new security risk: adversaries can embed malicious objectives in cross-application mobile content and steer agent behavior through indirect multimodal injection. Although prompt injection has been studied in desktop and web settings, its cross-modal…

Read at arXiv (research papers)

arXiv (research papers) Zesheng Shi SI

Identify, Classify, Act: Per-Instruction Controlled Compliance for LLMs over External Data

LLMs over external data face a per-instruction tension: a single document or tool return often interleaves legitimate directives with adversarial prompt injections, and the model must follow the former while rejecting the latter. Existing defenses suppress or trust all external instructions, collapsing the safety-utility frontier and forcing a choice between over-refusal and over-compliance. We argue the missing capability is \textbf{per-instruction credit attribution} and…

Read at arXiv (research papers)

arXiv (research papers) L. Zhu SI

Stealthy Poisoning Attacks via Reused KV Cache of Reasoning

Modern large language model (LLM) systems reuse key-value (KV) caches to accelerate inference. Recent work has shown that reusing KV caches from reasoning traces can also improve task performance, similar to retrieval-augmented generation (RAG). We demonstrate that this form of KV cache reuse introduces new poisoning vulnerabilities. We propose R$^2$ECAP (Reused Reasoning-Cache Poisoning), a black-box attack that plants poisoned reasoning traces solely through interactions…

Read at arXiv (research papers)

Journal articles (OpenAlex) Trung Thanh Bui · Asian-Pacific Economic Literature Geopolitics

Monetary Policy in Vietnam: Changes Over Time

ABSTRACT In the past decades, Vietnam has experienced a prolonged period of rapid economic growth. Many argued that the State Bank of Vietnam (SBV), which is an essential part of the Vietnamese government, makes a significant contribution to that achievement. However, high inflation in certain periods raises doubts about that compliment. Furthermore, studies of monetary policy in Vietnam mostly relied on constant‐parameter models such as VAR, which may not be proper as…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Moh. Dulkiah, Ahmad Ali Nurdin, Akmaliyah Akmaliyah et al. · Discover Global Society Geopolitics

Differentiating Islamic values and their associations with Muslim community participation and rural community development

This study examines how ukhuwah (brotherhood), amanah (trust and responsibility), and broader Islamic values influence Muslim community participation and rural community development in Indonesia. Using survey data from 693 Muslim middle-class respondents in rural areas of West Java, Central Java, and Yogyakarta, the study applies Partial Least Squares Structural Equation Modeling (PLS-SEM). The results show that Islamic values strongly influence Muslim community…

Read at Journal articles (OpenAlex)

MIT News AI Shafaq Zia | Whitehead Institute SI

AI model decodes the language cells use to communicate

A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.

Read at MIT News AI

arXiv (research papers) Fardin Ahsan Sakib SI

The Model Knows Better: Defending Medical Multi-Agent Systems Against Prompt Injection

Medical multi-agent systems, in which LLM agents discuss a case, are increasingly studied for clinical decision support. Prompt injection, text that an attacker adds to a model's input, can manipulate LLMs into wrong decisions. How it affects these systems, and how to protect them, remains unclear. Across two systems, two model families and three benchmarks, a single injected passage succeeds in up to 61% of cases, and collaboration does not consistently protect against it…

Read at arXiv (research papers)

Just Security Bulelani Jili Geopolitics

The Mythos of Digital Sovereignty

AI models such as Anthropic's Claude Mythos raise questions about access to frontier AI and institutional capacity.

Read at Just Security

Older stories