small language model

OSIRIS v4.5.3: a governed, local-first console for testing whether a small language model learns from conversation

v4.5.3 is a repair release. The exchange ledger's writer now chains each entry to the ledger file's last entry under a lock, so two consoles can no longer fork the hash chain; the chain check reports forks separately from damage (the author's ledger: 123 entries intact, seven forks from 2026-10-07, nothing rewritten)....

Interval Spectra of Small Finite Posets: Four Invariants Through Seven Points and an Eight-Point Obstruction

Scientific summary For a finite partially ordered set (poset), the interval-size distribution counts comparable pairs according to the number of elements strictly between them. We prove that, for posets with at most seven elements and a fixed cardinality, this complete distribution is determined by four statistics: the...

Student prescription medicine abuse schema

# Student Prescription Database A university database project on prescription patterns related to non-medical study-drug use among students. It combines a normalized prescription model, fictional teaching records and real FDA product and recall data. ## Purpose and scope Our societal topic is study-drug use and academi...

Fine-Tuning Small Language Models For Valid Universal Variability Language Generation

Few-Shot Learning for Low-Resource Natural Language Inference for French

Archetype analysis of multiplexed antibody imaging reveals interpretable morphological programs in optical pooled screens

High-content imaging-based screens with small molecule or genetic perturbations rely on rich optical phenotypes, yet how distinct subcellular markers contribute complementary biological information remains unclear. Here, we introduce CellMAPP, a multiplexed organelle imaging platform that uses optical pooled screens wi...

Occupational Health Risks Among Abattoir Workers in Sub-Saharan Africa: A One Health Systematic Review and Meta-Analysis Protocol

Abattoir workers in sub-Saharan Africa work where human, animal and environmental health meet. They are in daily contact with live animals, carcasses, blood, offal, hides and contaminated water. This exposes them to zoonotic infections such as brucellosis, Q fever, bovine tuberculosis and leptospirosis. They also face...

Evaluating Synthetic Survey Responses with Longitudinal Panel Data

Interest in using large language models to generate synthetic survey responses has grown as researchers seek faster, scalable approaches to survey data collection. However, less is known about whether longitudinal probability-based panels improve synthetic-response predictions and whether improvements in predictive acc...

Hybrid Transformer Architecture for Context-Aware Spam Email Classification

Spam and phishing email remain a persistent cybersecurity threat and a primary delivery vector for credential theft, malware, and ransomware. Detection accuracies above 99% are routinely reported. Such figures are typically obtained on small, single-source, monolingual corpora in which near-duplicate campaign templates...

Pragmatic AI: building low-impact LLM tools for energy research impact

Extended abstract 3-063-26 Large language models (LLMs) provide a powerful means to search for and summarise content. However, their training and use is a rapidly growing driver of energy and resource use. In this presentation we describe the development of an LLM-based search tool for the UKRI Energy Demand Research C...

task-allocation-runtime: allocating decisions across people, language models and deterministic tools

Software and Study 1 records for the article “Allocating cognitive work across people, language models and deterministic tools: A task-analytic method” (Simon Vajda, Applied Artificial Intelligence Institute, Deakin University). The runtime allocates the decisions in a cognitive task among deterministic tools, small la...

Retrieval‐Augmented Lesson Generation With Large Language Models for Scenario‐Based Tutor Training

High‐impact tutoring can improve student learning, but scaling tutoring programmes requires large numbers of novice tutors. Developing high‐quality tutor‐training lessons remains a key bottleneck. This study compared post‐test performance after participants studied human‐created or LLM‐created tutor‐traini...