Year: 2025
Type: conference-paper
Abstract: Biomedical texts, such as research articles and clinical reports, are often written in highly technical language, making them difficult for patients and the general public to understand.The BioLaySumm 2025 Shared Task addresses this challenge by promoting the development of models that generate lay summarisations of biomedical content.This paper focuses on Subtask 2.1: Radiology Report Generation with Layman's Terms.In this work, we evaluate two large language model (LLM) architectures, T5-large (700M parameter encoder-decoder model) and LLaMA-3.2-3B(3B parameter decoder-only model).Both models are trained under fully-supervised conditions using the task's multi-source dataset.Our results show that T5-large consistently outperforms LLaMA-3.2-3Bacross nine out of ten metrics, including relevance, readability, and clinical accuracy, despite having only a quarter of the parameters.Our T5-based model achieved the top rank in both the open-source and closesource tracks of the subtask 2.1.
Authors: Wenjun Zhang, Shekhar Chandra, Bevan Koopman, Jason Dowling, Aaron Nicolson
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4412889149