Artifact for: Evaluating LLMs in the Context of a Functional Programming Course: A Comprehensive Study

This artifact accompanies the paper “Evaluating LLMs in the Context of a Functional Programming Course: A Comprehensive Study” The artifact primarily provides the benchmark tasks, model outputs, and analysis scripts used in the study. It is intended to support verification of the empirical claims reported in the paper, rather than full end-to-end reproducibility. Due to the evolving nature of large language models and the use of the LearnOCaml platform for task creation and compilation, re-executing the complete experimental pipeline may not be feasible for all users. Instead, the artifact enables independent verification by allowing users to reproduce the figures and quantitative results reported in the paper using the provided scripts and data. Detailed explanations of the benchmark makeup and data organization are included in the accompanying documentation.