Conference Paper

An End-to-End Approach to Fine-Tune Small LLMs for Generating Admiral R Code in Statistical Programming

Jaime Yan, Tingting Tian

PhUSE US Connect 2025 · 2025 · OS08

Abstract

Base small LLMs perform poorly on Admiral code generation (OPS 0.36), and even strong general-purpose LLMs offer no built-in correctness verification for ADaM compliance. This paper presents an end-to-end pipeline for fine-tuning small, locally deployable LLMs on curated Admiral examples — fine-tuning LLaMA 3.1 8B with LoRA and validating output through a knowledge-graph system built from Admiral documentation. The pipeline covers data preparation, training, and evaluation, achieving reliable ADaM code generation without sending proprietary data to external APIs.

Keywords

fine-tuning · small LLMs · admiral · R · ADaM · LoRA · knowledge graph

Cite

Jaime Yan, Tingting Tian. "An End-to-End Approach to Fine-Tune Small LLMs for Generating Admiral R Code in Statistical Programming." PhUSE US Connect 2025, 2025. doi:10.5281/zenodo.22182891.