Research & development

Research Library

Understand the work. Examine the evidence. Build on what we learn.

Explore HelixScribe research into AI reliability, prompt behaviour and practical technology. Each record connects an accessible explanation with the published paper, methods, findings and supporting materials.

Search the research library

2 research papers

AI reliability research

Preprint · Version 3HSRN-2025-01

Operator-Level Prompting as Soft Behavioural Control in LLMs: Evidence from a 7.4× Manifold Compression

Claire Nicholson · 2 September 2026

This study compares conventional prompting with the same tasks wrapped in a structured operator format. It examines whether the resulting outputs vary less in their shape, language and adherence to instructions, without changing model weights. The paper reports a narrower range of output behaviour in the tested conditions; this is not a demonstration that outputs are universally correct, safe or deterministic.

DOI: 10.5281/zenodo.22260675

Explore the research

AI reliability research

Preprint · Version 1arXiv:2601.19934

Quantifying non-deterministic drift in large language models

Claire Nicholson · 12 January 2026

How repeatable are LLM responses when no stabilising prompt technique is applied? This study compares repeated outputs from two models across prompt categories, prompting modes and temperatures. It reports variation even at temperature 0.0, providing a baseline for later work on behavioural control. Its measures concern wording and output variation, rather than factual accuracy or the success of a mitigation technique.

DOI: 10.48550/arXiv.2601.19934

Explore the research

Research and accessible explanations

Paper records document the original investigations. Our explanations provide a starting point for understanding the concepts.

Read: Why LLMs drift →