Fine-Tuning: An Approach to Improve the Validity and Reliability of Large Language Models in Narrative Policy Framework Coding

• In Event: New Frontiers in the Narrative Policy Framework

Brief Overview

This study reviews ICR results from the NPF literature to contextualize recent LLM NPF coding efforts. Then a fine-tuning method is proposed and evaluated on a novel dataset, the resulting open-source model is presented with its interface.

Authors

• Nathan Jeschke, University of Tennessee, Knoxville

• Sean Liner, University of Tennessee, Knoxville