Galactic ChitChat: Using Large Language Models to Converse with Astronomy Literature

TL;DR

This paper explores using GPT-4 to interact with astronomy literature through in-context prompting and document distillation, demonstrating the model's ability to provide detailed, context-aware responses across multiple papers.

Contribution

It introduces a method for distilling astronomy papers to improve LLM interaction and evaluates GPT-4's effectiveness in multi-document comprehension within astronomy research.

Findings

GPT-4 provides detailed, context-aware answers to astronomy papers.

Distillation reduces input size by 50% while preserving semantic integrity.

GPT-4 excels in multi-document understanding for astronomy literature.

Abstract

We demonstrate the potential of the state-of-the-art OpenAI GPT-4 large language model to engage in meaningful interactions with Astronomy papers using in-context prompting. To optimize for efficiency, we employ a distillation technique that effectively reduces the size of the original input paper by 50\%, while maintaining the paragraph structure and overall semantic integrity. We then explore the model's responses using a multi-document context (ten distilled documents). Our findings indicate that GPT-4 excels in the multi-document domain, providing detailed answers contextualized within the framework of related research findings. Our results showcase the potential of large language models for the astronomical community, offering a promising avenue for further exploration, particularly the possibility of utilizing the models for hypothesis generation.