small language model

Comprehension as Compliance: How Grounded Content, Not Form, Moves Small Language Models to Act, and Where It Does Not

A small language model, given a task prefixed with a description of a recurring decision, will often recognize the decision and fail to act on it. This paper draws together a program of studies on that effect and states what drives it. The effect is carried by the content of the description, not by its form: a plain im...

The Martian Federalist Series 1: On the Necessity and Principles of Martian Confederation

Abstract The Martian Federalist Series 1 is a twelve-essay series (Numbers 1–12) explaining and defending the proposed Charter for the United Martian States—a voluntary confederation of sovereign city-states designed for the unique realities of early Mars settlement. The series addresses the inhabitants of Martian habi...

The Charter of the United Martian States: A Proposed Constitution for the Future Settlements on Mars

Abstract This document presents a complete constitutional framework for a future human society on Mars. The Charter of the United Martian States is a confederative instrument designed for a small, dispersed, high-risk frontier environment in which sovereign constituent polities voluntarily cooperate under strict limits...

An Interactive Computational Tool Using the Gauss-Seidel Method for Power Flow Analysis in Electrical Engineering Education

The development and implementation of an interactive computational tool in MATLAB for power flow analysis is presented. The software addresses the challenges of modeling electric power systems under complex topologies and incomplete data. By applying the iterative Gauss-Seidel method, which optimizes computational load...

Paired neuronal-haemodynamic neurovascular coupling measures and cognitive performance in vascular pathology: protocol for a systematic review and meta-analysis

This review will systematically synthesise existing findings, identify remaining knowledge gaps, and outline priorities for future research, to evaluate the association between paired neuronal–haemodynamic NVC measures and cognitive performance in middle-aged and older adults with vascular pathology.

LLM Compression with Jointly Optimizing Architectural and Quantization Choices

Retina as a window to systemic hypertension: From hypertensive retinopathy to artificial intelligence-driven microvascular biomarkers

The retinal microvasculature is the only vascular bed in the human body that can be visualised directly, non-invasively and repeatedly in vivo.Because retinal arterioles share embryological origin, calibre and autoregulatory physiology with the cerebral and coronary microcirculation, hypertension-induced retinal change...

Parallelism, critical windows, and separations among diffusion language models

A fine-grained comparison of the capacity for parallelism among these three leading approaches is initiated and it is proved that uniform and Gaussian diffusion can sample in a number of forward passes which scales with the dual total correlation of the underlying distribution.

Beyond Shadow Weights: Quantization-Aware Training as Quantized-Endpoint Descent

An idealized model for the residual phase is proposed, which records where each shadow weight sits inside its quantization cell as a fraction of the cell width as a fraction of the cell width, which leads to a crossing law that determines which coordinates cross quantization boundaries after a shadow update.

Labelling Bug-Fixing Commits with Local Open-Weight Language Models

This paper examines whether commits can be labelled as bug fixes from their content alone, using open-weight language models that run locally and therefore keep the process reproducible, inexpensive at corpus scale, usable on proprietary code, and independent of any issue tracker.

Two Fault Lines: Latent Polarity Geometry in X Community Notes

It is argued for a bridging model with more than one axis of disagreement, and for recruiting raters in the languages the current design reaches least, for recruiting raters in the languages the current design reaches least.

What Stops a Small Language Model From Driving a Database Agent

Small open-weight language models are assumed to fail at agentic database work because they lack the reasoning capacity for it, and a test against a production system finds that that assumption is false.