Weighting What Matters: Boosting Sample Efficiency in Medical Report Generation via Token Reweighting

Alexander Weers, Daniel Rueckert, Martin J. Menten

Keywords: Vision-Language Models, Sample Efficiency, Ophthalmology

TL;DR: Token weighting based on clinical keywords drastically improves sample efficiency in vision-language models for medical report generation.

Registration Requirement: Yes

Abstract:

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.

Visa & Travel: No

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 129