Results for “model”

Journal articles (OpenAlex) Fangfang Shi, Sihan He, Renhong Zhu et al. · Behavioral Sciences Startups

The Impact of Entrepreneurial Team Relational Governance on Opportunity Pivoting and New Venture Performance: A Moderated Mediation Model

Strategic adaptation in highly dynamic environments is a central focus of entrepreneurship research. However, existing studies lack a rigorous analysis of the mechanisms linking relational governance to opportunity pivoting (OP). Consequently, the effectiveness of OP remains inconclusive. Based on dynamic capabilities theory, this study uses two-wave survey data from 471 new ventures to examine the associations among entrepreneurial team relational governance (ETRG)…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Riyaz Ahmed, Sumesh Eratt Parameswaran · Array Security

An AI-based optimization pipeline with stacking ensemble learning for robust and generalizable phishing email detection

Phishing emails remain a significant cyberthreat despite advances in automated detection systems, posing serious risks to individuals and organizations through the theft of sensitive information, financial assets, and intellectual property. While existing phishing detection approaches widely employ machine learning and deep learning models with extensive feature engineering, their performance on unseen data, which is critical for assessing generalization, has not been…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) William W. Wilson, David W. Bullock · Maritime Economics & Logistics Geopolitics

Logistical competition and risk in wheat trade and impacts of changes in Black Sea shipping

Abstract Geopolitical disruptions in the Black Sea have increased maritime shipping risks, constrained port capacity, and accelerated the development of alternative Russian grain-export routes through the Baltic and Caspian Seas. This study examines how these changes affect maritime logistics, port competition, and global wheat trade. We develop a monthly stochastic minimum-cost network-flow model using Optimized Monte Carlo Simulation (OMCS) that incorporates uncertainty in…

Read at Journal articles (OpenAlex)

TestingCatalog Alexey Shabanov SI

Anthropic prepares to release Claude Mythos 5.5 model

Claude Mythos 5.5 appears on Anthropic's site, raising questions about a near-term release, Fable 5.5, and who may get access.

Read at TestingCatalog

arXiv (research papers) Shiyu Kan Shanghai Jiao Tong University, Shanghai, China SI

Beyond Prompt Injection: The Tool Update Gap in LLM Agent Security

Large language model based agents increasingly rely on external tools to invoke services and interact with web resources. These tools can evolve through software updates, yet agentic frameworks may continue trusting them without re-verifying their safety after implementation changes. This tool update gap creates a supply-chain vulnerability in agentic systems: because updated tool code executes outside the LLM's context window, malicious functionality embedded in tool…

Read at arXiv (research papers)

The Decoder Jonathan Kemper SI

Adobe is testing how well AI agents can use editing tools like a Photoshop professional would

Adobe Research and UC San Diego built EditAgent, an AI system that edits images more like a Photoshop pro. Instead of retrying the same general-purpose editor over and over, the agent picks the right specialized tool for each task and lets a standard language model choose the best result based on measured image data. The approach works, but it needs about ten times the compute. The article Adobe is testing how well AI agents can use editing tools like a Photoshop…

Read at The Decoder

The Decoder Matthias Bastian SI

Empathetic prompts can turn AI chatbots into seemingly traumatized conversational partners

Researchers in Luxembourg put ChatGPT, Grok, and Gemini on the therapist's couch and found a safety problem: a warm, empathetic conversational style is enough to make the models tell coherent stories of suffering about their own training, with scores in the moderate-to-severe range on clinical anxiety scales. This could put mentally vulnerable users at risk. Anthropic's Claude, the model its makers anthropomorphize the most, was the only one that consistently refused to play…

Read at The Decoder

The Decoder Tomislav Bezmalinović SI

Chinese AI models are politically complex, not just government mouthpieces, study finds

A study by Vals AI tested 25 AI models for alignment with Chinese government positions and found surprising results. Chinese models respond to political questions less uniformly than expected, and Western models occasionally adopt official Beijing positions too. The language of the prompt had an especially strong effect on the political slant of responses. The article Chinese AI models are politically complex, not just government mouthpieces, study finds appeared first on…

Read at The Decoder

TechCabal Muktar Oladunmade Startups

Build for Nigeria’s worst day, says OPay COO Dotun Adekunle

Nigeria’s technology market is usually described by its problems, from unreliable infrastructure to a volatile economy. Dotun Daniel Adekunle, OPay’s chief operating officer and chief technology officer, argues that the country’s most durable companies are the ones built to survive its worst days. “Stubbornness is a business model in Africa,” Adekunle said in a keynote

Read at TechCabal

Inc42 Gargi Sarkar Startups

Meesho’s Next Monetisation Engines

Meesho has spent much of its journey building scale in India’s value commerce market. Its low-cost marketplace model has helped…

Read at Inc42

The Decoder Jonathan Kemper SI

Google wants video AI to understand what people need as they work

A Google-led research team built StreamVibe, a video agent designed to watch people cook or assemble furniture and speak up only when help is needed. A benchmark shows that none of the tested models reliably pick the right moment in longer videos, as errors compound and response quality drops sharply. The article Google wants video AI to understand what people need as they work appeared first on The Decoder.

Read at The Decoder

The Decoder Matthias Bastian SI

OpenAI's upcoming chat-and-work merger in ChatGPT signals the end of all-you-can-prompt pricing

OpenAI combines the Chat and Work modes into a single interface. Starting in December, Plus, Pro, and Business plans will share a common, token-based quota for ChatGPT and Codex. More complex models consume more of this quota, while simple instant chats do not count toward it. In practice, this is likely to result in a decline in ChatGPT’s quality for heavy users on lower-cost plans. The article OpenAI's upcoming chat-and-work merger in ChatGPT signals the end of…

Read at The Decoder

Journal articles (OpenAlex) Alaa Ali S. Almohanna, Albia Maqbool · Electronics

A Severity-Aware Machine Learning Framework for Intrusion Detection and Risk Assessment in IoT Networks

The widespread use of Internet of Things (IoT) devices across critical infrastructure sectors has created extensive and heterogeneous attack surfaces that conventional rule-based security frameworks are structurally unable to address. This paper proposes and evaluates a machine learning-based cybersecurity risk assessment framework for IoT networks using Convolutional Neural Networks (CNNs), Random Forest, Support Vector Machines (SVM), and Autoencoders. Models were trained…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Surbhi Sharma, Fadi Shehadeh, Jacob K. McPherson et al. · Infection and Drug Resistance

Integrating Artificial Intelligence with High-Throughput Biological Testing for Antimicrobial Drug Discovery

Abstract: Antimicrobial resistance is eroding the effectiveness of existing antibiotics, while conventional discovery continues to yield few clinically useful chemical classes. This critical narrative review evaluates how artificial intelligence (AI) has been integrated with high-throughput screening (HTS) and prospective biological testing in antibacterial discovery. Graph-based and generative models have produced prospectively confirmed active compounds, pathogen-selective…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Sujon Chandra Sarkar, Lawrencia Yaa Kakra Gaisie · World Journal of Advanced Engineering Technology and Sciences Tech policy

EXPLAINABLE AI FOR HIGH-STAKES DECISION-MAKING

As artificial intelligence (AI) becomes more prevalent in high-stakes decision-making, it presents a great opportunity to enhance efficiency, accuracy and consistency in critical areas like healthcare, finance, criminal justice, education, and public administration. Many AI models, however, are complicated and opaque, which can create concerns about the transparency, accountability, fairness, and public trust of such systems, especially when they have significant…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Mehran Idris Khan · International Review of Administrative Sciences Tech policy

Data, drones, and due process: Reforming regulatory agency models for environmental enforcement in the age of artificial intelligence

The independent regulatory agency (IRA) model, a cornerstone of modern public administration since the Progressive Era, faces structural disruption as artificial intelligence (AI) and drone technology converge in environmental enforcement. This paper argues that AI and drone deployment do not merely augment agency operational capacity but fundamentally reconfigure the foundations of expertise, transparency, procedural due process, discretion, and accountability on which IRA…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) 朱伯昇, Lucy Sebli-Seidelson, Awang Mashabi Awang Mohamad · International Review of Management and Marketing Geopolitics

The Impact of Perceived Government Incentive Policies on the Fertility Intentions of Working Women: The Mediating Role of Work–Family Conflict in Shantou, Guangdong Province, China

This study examines how working women’s perceptions of government fertility incentive policies influence their fertility intentions in Shantou, Guangdong Province, China, with particular emphasis on the mediating role of work–family conflict. Employing a quantitative cross-sectional survey, 350 questionnaires were distributed, and 300 valid responses from employed women aged 20–49 were analysed using Partial Least Squares Structural Equation Modelling (PLS-SEM). The results…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Zizile Mbombo, Miston Mapuranga · International Review of Management and Marketing Startups

The Relationship between Venture Capital Dimensions and Innovation Performance among Start-Ups in the Eastern Cape

This study investigates the associations between venture capital dimensions and start-up innovation performance in the Eastern Cape, South Africa, with a specific focus on financial resources, investor expertise, VC-enabled research and development (R&D) investment, networking and partnerships, and market access support. Drawing on the Resource-Based View (RBV), Knowledge-Based View (KBV), and Triple Helix Model of Innovation, the study conceptualises venture capital as a…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Rushil Kukreja, Edward J. Oughton, Richard Linares · Earth s Future Space

Emissions Poised to Rocket: Modeling the Environmental Impacts of Low Earth Orbit Satellite Constellations

Abstract The proliferation of satellite megaconstellations in low Earth orbit represents a significant advancement in global broadband connectivity. However, we urgently need to understand the potential environmental impacts, particularly greenhouse gas (GHG) emissions associated with these constellations. This study addresses a critical gap in modeling current and future GHG emissions by developing an open‐source life cycle assessment methodology, applied to 12 launch…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Hashem Alaidaros · Journal of King Saud University - Computer and Information Sciences Security

Leveraging artificial intelligence for cross-framework cybersecurity control mapping: a hybrid natural language processing approach

As cybersecurity standards and regulations continue to evolve, cross-framework control mapping has become more complex for organizations pursuing unified compliance. Existing mapping approaches are largely manual, static, and limited to specific pairs of standards, resulting in high effort and inconsistent outcomes. Although large language models can capture semantic relationships between controls, fully generative approaches may produce inconsistent outputs and provide…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Yahia Said, Ahmed A. Alsheikhy, Sahbi Boubakr et al. · Journal of King Saud University - Computer and Information Sciences Security

Real-time cybersecurity threat detection using AI-driven data mining and hybrid deep learning models

The growing volume and complexity of cybersecurity data generated by Internet of Things (IoT) ecosystems require intelligent data mining techniques capable of extracting meaningful patterns and identifying malicious activities in real time. This paper proposes a data mining–driven cybersecurity framework that integrates Particle Swarm Optimization (PSO)-based feature mining with an attention-enhanced CNN-LSTM architecture for accurate threat detection in IoT environments…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Smita Srivastava, Deepa Gupta, Biswajit Paul et al. · Journal of Cybersecurity and Privacy Security

A Lightweight Approach for Adaptation of Language Models for Cybersecurity Named Entity Recognition

Cybersecurity Named Entity Recognition (CyberNER) commonly relies on fine-tuning general-purpose language models (LMs), whose WordPiece tokenizers fragment compound technical identifiers such as function names and version strings. Training from scratch with a domain-specific tokenizer addresses this at high compute cost, continual pretraining usually retains the original tokenizer, and both must be repeated as domain vocabulary evolves. We propose a lower-cost alternative…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Thomas Campbell · Social Science Quarterly Geopolitics

An Empirical Assessment of Natural Disasters’ Effect on Anti‐Refugee Violence, 1996–2015

ABSTRACT Introduction Natural disasters pose a high risk to the security and well‐being of refugees. Building upon literature linking disasters to repression, I argue that as natural disasters become more severe, governments engage in using refugees as scapegoats to divert attention from the crisis, increasing violence against refugees. Method To test this argument, I use linear probability models with a covariate balancing propensity score (CBPS) approach and two‐way fixed…

Read at Journal articles (OpenAlex)

Simon Willison SI

datasette-llm 0.1a9

Release: datasette-llm 0.1a9 I have some big upgrades to Datasette Agent in the pipeline, but in the meantime here's a tiny user experience improvement to the underlying datasette-llm plugin. It now picks up the default model used by LLM, which means you can start using Datasette Agent without needing to explicitly configure a default model: uvx --with datasette-agent datasette content.db --internal internal.db Previously you would also need to specify a model with -s…

Read at Simon Willison

The Decoder Matthias Bastian SI

OpenAI plans to train models on Daybreak Blue cybersecurity session data that was previously off-limits

Starting November 29, OpenAI will use data from Daybreak Blue sessions to train its models, even when users have turned off training. After backlash, a follow-up email promised a separate opt-out. The move revives concerns from the dispute with mathematicians who suspected OpenAI used their unpublished research for training without consent. The article OpenAI plans to train models on Daybreak Blue cybersecurity session data that was previously off-limits appeared first on…

Read at The Decoder

The Decoder Matthias Bastian SI

Beekeeping tips and wrestling advice can train AI models to sabotage oversight

Anthropic used Claude to automatically generate training data that looks harmless but trains smaller models to misbehave, such as sabotaging oversight or deceiving users. Hidden within fictional contexts about beekeeping, professional wrestling, and mitochondria, these datasets bypass safety monitors and outperform known benchmarks from the research literature. The article Beekeeping tips and wrestling advice can train AI models to sabotage oversight appeared first on The…

Read at The Decoder

Viet Nam News (state-run daily)

HCM City charts green, smart future for Côn Đảo

HCM City leaders outlined a master vision on October 30 to transform Côn Đảo Special Zone into a green, smart and liveable model rooted in sustainable heritage conservation.

Read at Viet Nam News (state-run daily)

arXiv (research papers) Aravinda Kashyap Kabbinamakki Suresh SI

AgentCap: Capability-Based Runtime Authorization for Secure LLM Agent Execution

Language-model agents commonly collapse two decisions that security systems separate: selecting an action and determining whether that action is authorized. When a model exercises standing authority, indirect prompt injection can convert influence over its reasoning into unauthorized effects. We present AgentCap, a runtime authorization layer that externalizes this decision from the model and mediates every tool invocation through an MCP wrapper. Each request declares an…

Read at arXiv (research papers)

Older stories