Leveraging Large Language Models for Zero-shot Lay Summarisation in Biomedicine and Beyond

TL;DR

This paper introduces a two-stage framework for zero-shot lay summarisation using Large Language Models, demonstrating improved human preference, the potential for LLMs as judges, and initial success in summarising NLP articles.

Contribution

The paper presents a novel two-stage framework for zero-shot lay summarisation, assesses LLMs as preference judges, and explores summarisation of NLP articles, advancing practical applications in biomedical and general domains.

Findings

Summaries are increasingly preferred by human judges for larger models.

LLMs can replicate human preferences as judges.

LLMs can generalise to summarising NLP articles.

Abstract

In this work, we explore the application of Large Language Models to zero-shot Lay Summarisation. We propose a novel two-stage framework for Lay Summarisation based on real-life processes, and find that summaries generated with this method are increasingly preferred by human judges for larger models. To help establish best practices for employing LLMs in zero-shot settings, we also assess the ability of LLMs as judges, finding that they are able to replicate the preferences of human judges. Finally, we take the initial steps towards Lay Summarisation for Natural Language Processing (NLP) articles, finding that LLMs are able to generalise to this new domain, and further highlighting the greater utility of summaries generated by our proposed approach via an in-depth human evaluation.