Saved in:
Bibliographic Details
Main Authors: Weers, Alexander, Rueckert, Daniel, Menten, Martin J.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.21082
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910158937915392
author Weers, Alexander
Rueckert, Daniel
Menten, Martin J.
author_facet Weers, Alexander
Rueckert, Daniel
Menten, Martin J.
contents Training vision-language models (VLMs) for medical report generation is often hindered by the scarcity of high-quality annotated data. This work evaluates the use of a weighted loss function to improve data efficiency. Compared to standard cross-entropy loss, which treats all token prediction errors equally, the reweighted loss shifts the focus to semantically salient tokens with outsized clinical importance. In experiments on ophthalmological report generation, we show that this simple method improves efficiency across multiple data scales, achieving similar report quality with up to ten times less training data.
format Preprint
id arxiv_https___arxiv_org_abs_2604_21082
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Weighting What Matters: Boosting Sample Efficiency in Medical Report Generation via Token Reweighting
Weers, Alexander
Rueckert, Daniel
Menten, Martin J.
Computation and Language
Machine Learning
Training vision-language models (VLMs) for medical report generation is often hindered by the scarcity of high-quality annotated data. This work evaluates the use of a weighted loss function to improve data efficiency. Compared to standard cross-entropy loss, which treats all token prediction errors equally, the reweighted loss shifts the focus to semantically salient tokens with outsized clinical importance. In experiments on ophthalmological report generation, we show that this simple method improves efficiency across multiple data scales, achieving similar report quality with up to ten times less training data.
title Weighting What Matters: Boosting Sample Efficiency in Medical Report Generation via Token Reweighting
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2604.21082