Research

Paper

TESTING March 16, 2026

Exemplar Diffusion: Improving Medical Object Detection with Opportunistic Labels

Authors

Victor Wåhlstrand, Jennifer Alvén, Ida Häggström

Abstract

We present a framework to take advantage of existing labels at inference, called \textit{exemplars}, in order to improve the performance of object detection in medical images. The method, \textit{exemplar diffusion}, leverages existing diffusion methods for object detection to enable a training-free approach to adding information of known bounding boxes at test time. We demonstrate that for medical image datasets with clear spatial structure, the method yields an across-the-board increase in average precision and recall, and a robustness to exemplar quality, enabling non-expert annotation. Moreover, we demonstrate how our method may also be used to quantify predictive uncertainty in diffusion detection methods. Source code and data splits openly available online: https://github.com/waahlstrand/ExemplarDiffusion

Metadata

arXiv ID: 2603.15267
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-16
Fetched: 2026-03-17 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.15267v1</id>\n    <title>Exemplar Diffusion: Improving Medical Object Detection with Opportunistic Labels</title>\n    <updated>2026-03-16T13:36:23Z</updated>\n    <link href='https://arxiv.org/abs/2603.15267v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.15267v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>We present a framework to take advantage of existing labels at inference, called \\textit{exemplars}, in order to improve the performance of object detection in medical images. The method, \\textit{exemplar diffusion}, leverages existing diffusion methods for object detection to enable a training-free approach to adding information of known bounding boxes at test time. We demonstrate that for medical image datasets with clear spatial structure, the method yields an across-the-board increase in average precision and recall, and a robustness to exemplar quality, enabling non-expert annotation. Moreover, we demonstrate how our method may also be used to quantify predictive uncertainty in diffusion detection methods. Source code and data splits openly available online: https://github.com/waahlstrand/ExemplarDiffusion</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-16T13:36:23Z</published>\n    <arxiv:comment>Submitted to MICCAI 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Victor Wåhlstrand</name>\n    </author>\n    <author>\n      <name>Jennifer Alvén</name>\n    </author>\n    <author>\n      <name>Ida Häggström</name>\n    </author>\n  </entry>"
}