Saved in:
Bibliographic Details
Main Authors: Grundmann, Paul, Fast, Dennis, Frick, Jan, Steffek, Thomas, Gers, Felix, Nejdl, Wolfgang, Löser, Alexander
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.26136
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915754446684160
author Grundmann, Paul
Fast, Dennis
Frick, Jan
Steffek, Thomas
Gers, Felix
Nejdl, Wolfgang
Löser, Alexander
author_facet Grundmann, Paul
Fast, Dennis
Frick, Jan
Steffek, Thomas
Gers, Felix
Nejdl, Wolfgang
Löser, Alexander
contents With their growing capabilities, generative large language models (LLMs) are being increasingly investigated for complex medical tasks. However, their effectiveness in real-world clinical applications remains underexplored. To address this, we present CliniBench, the first benchmark that enables comparability of well-studied encoder-based classifiers and generative LLMs for discharge diagnosis prediction from admission notes in MIMIC-IV dataset. Our extensive study compares 12 generative LLMs and 3 encoder-based classifiers and demonstrates that encoder-based classifiers consistently outperform generative models in diagnosis prediction. We assess several retrieval augmentation strategies for in-context learning from similar patients and find that they provide notable performance improvements for generative LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2509_26136
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CliniBench: A Clinical Outcome Prediction Benchmark for Generative and Encoder-Based Language Models
Grundmann, Paul
Fast, Dennis
Frick, Jan
Steffek, Thomas
Gers, Felix
Nejdl, Wolfgang
Löser, Alexander
Computation and Language
With their growing capabilities, generative large language models (LLMs) are being increasingly investigated for complex medical tasks. However, their effectiveness in real-world clinical applications remains underexplored. To address this, we present CliniBench, the first benchmark that enables comparability of well-studied encoder-based classifiers and generative LLMs for discharge diagnosis prediction from admission notes in MIMIC-IV dataset. Our extensive study compares 12 generative LLMs and 3 encoder-based classifiers and demonstrates that encoder-based classifiers consistently outperform generative models in diagnosis prediction. We assess several retrieval augmentation strategies for in-context learning from similar patients and find that they provide notable performance improvements for generative LLMs.
title CliniBench: A Clinical Outcome Prediction Benchmark for Generative and Encoder-Based Language Models
topic Computation and Language
url https://arxiv.org/abs/2509.26136