Designing an Evaluation Framework for Large Language Models in Astronomy Research

TL;DR

This paper proposes an evaluation framework for assessing how astronomy researchers interact with Large Language Models, using a Slack chatbot with retrieval-augmented generation grounded in arXiv papers, enabling future dynamic assessments.

Contribution

It introduces a novel experimental design and data collection method for evaluating LLMs in astronomy research contexts.

Findings

Deployment of a Slack chatbot for astronomy queries

Collection of user feedback and interaction data

Framework enables future dynamic evaluations

Abstract

Large Language Models (LLMs) are shifting how scientific research is done. It is imperative to understand how researchers interact with these models and how scientific sub-communities like astronomy might benefit from them. However, there is currently no standard for evaluating the use of LLMs in astronomy. Therefore, we present the experimental design for an evaluation study on how astronomy researchers interact with LLMs. We deploy a Slack chatbot that can answer queries from users via Retrieval-Augmented Generation (RAG); these responses are grounded in astronomy papers from arXiv. We record and anonymize user questions and chatbot answers, user upvotes and downvotes to LLM responses, user feedback to the LLM, and retrieved documents and similarity scores with the query. Our data collection method will enable future dynamic evaluations of LLM tools for astronomy.