In this paper, we consider the challenge of summarizing patients' medical progress notes in a limited data setting. For the Problem List Summarization (shared task 1A) at the BioNLP Workshop 2023, we demonstrate that Clinical-T5 fine-tuned to 765 medical clinic notes outperforms other extractive, abstractive and zero-shot baselines, yielding reasonable baseline systems for medical note summarization. Further, we introduce Hierarchical Ensemble of Summarization Models (HESM), consisting of token-level ensembles of diverse fine-tuned Clinical-T5 models, followed by Minimum Bayes Risk (MBR) decoding. Our HESM approach lead to a considerable summarization performance boost, and when evaluated on held-out challenge data achieved a ROUGE-L of 32.77, which was the best-performing system at the top of the shared task leaderboard.
@article{arxiv.2306.05317,
title = {CUED at ProbSum 2023: Hierarchical Ensemble of Summarization Models},
author = {Potsawee Manakul and Yassir Fathullah and Adian Liusie and Vyas Raina and Vatsal Raina and Mark Gales},
journal= {arXiv preprint arXiv:2306.05317},
year = {2023}
}