Comparative Analysis of Abstract vs Full-Text Biomedical Summarization

This study evaluates whether biomedical lay summarization requires full-text processing or if abstracts suffice for readability and factual consistency. It compares few-shot learning against task-specific fine-tuning across two recent models to optimize computational efficiency for patient-facing tools.