Category
3,240 papers
quickthink is a local-first inference control layer that aims to give small, on-device language models more reliable structured output by inserting a compressed plan-then-answer scaffold with latency-aware routing. Its routing, plan-grammar, and validation logic run as plain-Python middleware (no extra model of its own...
quickthink is a local-first inference control layer that aims to give small, on-device language models more reliable structured output by inserting a compressed plan-then-answer scaffold with latency-aware routing. Its routing, plan-grammar, and validation logic run as plain-Python middleware (no extra model of its own...
Agentic artificial intelligence (AI) is shifting from decision support to autonomous coordination, challenging established assumptions about managerial authority. This qualitative conceptual paper synthesises recent literature in algorithmic management, organisational theory, and human-computer interaction to examine h...
Agentic artificial intelligence (AI) is shifting from decision support to autonomous coordination, challenging established assumptions about managerial authority. This qualitative conceptual paper synthesises recent literature in algorithmic management, organisational theory, and human-computer interaction to examine h...
Reach audiences
Advertise in front of researchers, engineers, and readers.
This study characterises the methodological architecture that produces apparent ML superiority and translates that diagnosis into minimum standards, and proposes six minimum methodological standards - pre-specification, fair comparator design, robust external or temporal validation, calibration and decision-analytic re...
In children (2–59 months), oral amoxicillin significantly reduces treatment failure compared to no antibiotics for fast breathing pneumonia and is comparable to injectable antibiotics for chest indrawing pneumonia without danger signs.
Primary computational corpus for the manuscript "Large language models that near-perfectly evaluate products systematically recommend a worse familiar brand" (Affonso, Spears School of Business, Oklahoma State University). Contains 627,491 controlled product-recommendation trials across 30 frontier language models from...
Quantum-inspired (QI) optimizers are increasingly proposed as alternatives to classical gradient-based training, yet are frequently benchmarked through classical stand-ins rather than themselves. Across tabular, vision, and language tasks we evaluate three classes: direct implementations of published update rules (Spal...
Uncer-tainTree is presented, an end-to-end process that makes uncertainty a first-class concern of multi-model consistency analysis and includes interacting uncertainties from various sources and models.
Despite growing interest in prompt engineering, its influence on user experience—particularly in relation to cultural communication norms—remains largely unexplored. This controlled experiment examined how respectful versus directive prompting—a culturally meaningful contrast in Korean—affects Korean users’ psychologic...
ZENODO SUBMISSION — DOI: 10.5281/zenodo.22693244 --- COMPLETE METADATA & DESCRIPTION Basic Information Field ValueTitle The Navier–Stokes Equation and the Noor Ocean in Mainstream Science Terms: A Complete Translation of N-K Sciences into Conventional Physics LanguageAuthor Malik Muhammad UsmanORCID 0009-0004-3269-2819...
Small administrative units in higher education—continuing-education offices, language and culture centers, advising and tutoring services, and similar back-office functions—are routinely expected to deliver responsive, data-informed service with the budgets and staffing of a support desk. Enterprise customer relationsh...
A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.
Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security. The post Proof-based general-purpose computing with Magnetar appeared first on Microsoft Research.
Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.
StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics. The post Simulating student learning for better AI tutoring appeared first on Microsoft Research.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
