ReLay: Personalized LLM-Generated Plain-Language Summaries for Better Understanding, but at What Cost?

TL;DR

This paper introduces ReLay, a dataset and evaluation of personalized health plain-language summaries generated by LLMs, highlighting improvements in comprehension but also risks like bias reinforcement and hallucinations.

Contribution

ReLay is a novel dataset that enables evaluation of personalized LLM-generated summaries, revealing benefits and challenges of personalization in health communication.

Findings

Personalization improves comprehension and perceived quality.

Personalization increases risk of biases and hallucinations.

Trade-off exists between personalization benefits and safety concerns.

Abstract

Plain Language Summaries (PLS) aim to make research accessible to lay readers, but they are typically written in a one-size-fits-all style that ignores differences in readers' information needs and comprehension. In health contexts, this limitation is particularly important because misunderstanding scientific information can affect real-world decisions. Large language models (LLMs) offer new opportunities for personalizing PLS, but it remains unclear whether personalization helps, which strategies are most effective, and how to balance personalization with safety. We introduce ReLay, a dataset of 300 participant--PLS pairs from 50 lay participants in both static (expert-written) and interactive (LLM-personalized) settings. ReLay includes user characteristics, health information needs, information-seeking behavior, comprehension outcomes, interaction logs, and quality ratings. We use ReLay to evaluate five LLMs across two personalization methods. Personalization improves comprehension and perceived quality, but it also raises the risk of reinforcing user biases and introducing hallucinations, revealing a trade-off between personalization and safety. These findings highlight the need for personalization methods that are both effective and trustworthy for diverse lay audiences.