small language model

Perancangan Sistem Informasi Penjualan Berbasis Web di Gudang Durian UMKM Surya Durian Medan

Surya Durian Medan is a micro, small, and medium enterprise engaged in durian sales and durian derivative product management. Its operational activities include receiving durian from suppliers, storing products in the warehouse, distributing products to sales stalls, processing durian into peeled durian and durian fles...

An Evidence-Constrained RAG Learning Support System for Interdisciplinary Smart Water Engineering Education: Multi-Source Knowledge Governance and Reproducible Technical Evaluation

Smart water engineering is inherently interdisciplinary, requiring students to synthesize hydrological and hydraulic principles, observational data, domain-specific models, engineering standards, and computational tools to support evidence-based reasoning and decision-making. Yet, the knowledge required for such learni...

How Many Bits Does a Fact Need? Memory Load, Competing Alternatives and Weight Quantization in a Small Transformer

This work investigates how post-training weight quantization affects the knowledge memorized by a language model. To this end, we train a transformer with 1.1 million parameters on synthetic biographies, a setting in which the amount of information is known exactly, and we vary both the information load and the number...

Deploying LLM Inference on a Repurposed UMA APU: Transferable Lessons from a Vulkan-Only, 16 GB Edge Platform

Cost-driven interest in running large language models (LLMs) on non-mainstream silicon outpaces the maturity of the surrounding software stacks. This paper uses one such platform, the AMD BC-250 (a repurposed cryptocurrency-mining board with a GFX1013 "Cyan Skillfish" accelerated processing unit (APU), 16 GB of unified...

Application of large language models for assigning clear cell likelihood score v2.0 from free-text MRI reports: a feasibility study.

RATIONALE AND OBJECTIVES To evaluate the performance of four large language models (LLMs) for automated feature extraction and clear cell likelihood score version 2.0 (ccLS v2.0) assignment from free-text reports using a multi-step prompting strategy. MATERIALS AND METHODS This retrospective multicenter study include...

Counting What Inspection Misses: Auditing an LLM-Generated Quiz Bank for an Educational Game

LLMs make it cheap for a small team to write the assessment content of an educational game, and the result usually reads well. We audited the 500-item bilingual multiple-choice bank of a shipped educational RPG, generated by an LLM coding assistant, against a plan fixed before any audit model was run. Factually, the ba...

On-Device Multimodal Small Language Models for Privacy- Preserving Edge Intelligence Through Quantization Pruning Distillation and Energy-Efficient Inference

On-device multimodal small language models are emerging as a practical foundation for privacypreserving edge intelligence, enabling devices to interpret and generate language while processing images, audio, video, sensor streams, and contextual signals without continuous cloud dependence. Their deployment can reduce da...

Towards sustainable and energy-efficient agentic AI via small language models: A comprehensive survey

Creation and Collapse Mechanism: Forward and Reverse Pathways, Quantization, and the 2.8\,M_\odot Fundamental Black Hole Unit

This work presents a comprehensive, plain-language framework detailing the symmetrical relationship between the forward construction of matter and its gravitational reversal. The hypothesis outlines a five-part model spanning from fundamental fields to black hole formation: Part 1: Building (Forward): Matter is constru...

Corporate Criminal Liability in Spain: Rethinking Article 31 bis of the Spanish Penal Code

Abstract The introduction of corporate criminal liability into Spanish law through Organic Law 5/2010 and its substantial reform by Organic Law 1/2015 marked a decisive departure from the traditional maxim societas delinquere non potest. Article 31 bis of the Spanish Penal Code now permits criminal liability of legal p...

LUCID: A Training-Free Few-Shot Framework for Leaf Disease Diagnosis via Visual–Knowledge Evidence Fusion

The diagnosis of leaf diseases in medicinal and non-mainstream plants is often hindered by the scarcity of labeled samples, whereas existing deep learning methods generally rely on large-scale labeled datasets. Prototype-based few-shot methods are constrained by the limited number of support samples and fine-grained vi...

Enhancing hybrid information retrieval with hyperdimensional computing: a framework for efficient and robust knowledge-intensive NLP

Retrieval-Augmented Generation (RAG) has emerged as a practical framework for augmenting large language models (LLMs) with external knowledge sources. However, standard RAG pipelines often remain computationally intensive and exhibit sensitivity to input noise, particularly in real-world scenarios where queries can be...