Paper

{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.23953v1</id>\n    <title>VOLMO: Versatile and Open Large Models for Ophthalmology</title>\n    <updated>2026-03-25T05:25:10Z</updated>\n    <link href='https://arxiv.org/abs/2603.23953v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.23953v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Vision impairment affects millions globally, and early detection is critical to preventing irreversible vision loss. Ophthalmology workflows require clinicians to integrate medical images, structured clinical data, and free-text notes to determine disease severity and management, which is time-consuming and burdensome. Recent multimodal large language models (MLLMs) show promise, but existing general and medical MLLMs perform poorly in ophthalmology, and few ophthalmology-specific MLLMs are openly available. We present VOLMO (Versatile and Open Large Models for Ophthalmology), a model-agnostic, data-open framework for developing ophthalmology-specific MLLMs. VOLMO includes three stages: ophthalmology knowledge pretraining on 86,965 image-text pairs from 26,569 articles across 82 journals; domain task fine-tuning on 26,929 annotated instances spanning 12 eye conditions for disease screening and severity classification; and multi-step clinical reasoning on 913 patient case reports for assessment, planning, and follow-up care. Using this framework, we trained a compact 2B-parameter MLLM and compared it with strong baselines, including InternVL-2B, LLaVA-Med-7B, MedGemma-4B, MedGemma-27B, and RETFound. We evaluated these models on image description generation, disease screening and staging classification, and assessment-and-management generation, with additional manual review by two healthcare professionals and external validation on three independent cohorts for age-related macular degeneration and diabetic retinopathy. Across settings, VOLMO-2B consistently outperformed baselines, achieving stronger image description performance, an average F1 of 87.4% across 12 eye conditions, and higher scores in external validation.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.ET'/>\n    <published>2026-03-25T05:25:10Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Zhenyue Qin</name>\n    </author>\n    <author>\n      <name>Younjoon Chung</name>\n    </author>\n    <author>\n      <name>Elijah Lee</name>\n    </author>\n    <author>\n      <name>Wanyue Feng</name>\n    </author>\n    <author>\n      <name>Xuguang Ai</name>\n    </author>\n    <author>\n      <name>Serina Applebaum</name>\n    </author>\n    <author>\n      <name>Minjie Zou</name>\n    </author>\n    <author>\n      <name>Yang Liu</name>\n    </author>\n    <author>\n      <name>Pan Xiao</name>\n    </author>\n    <author>\n      <name>Mac Singer</name>\n    </author>\n    <author>\n      <name>Amisha Dave</name>\n    </author>\n    <author>\n      <name>Aidan Gilson</name>\n    </author>\n    <author>\n      <name>Tiarnan D. L. Keenan</name>\n    </author>\n    <author>\n      <name>Emily Y. Chew</name>\n    </author>\n    <author>\n      <name>Zhiyong Lu</name>\n    </author>\n    <author>\n      <name>Yih-Chung Tham</name>\n    </author>\n    <author>\n      <name>Ron Adelman</name>\n    </author>\n    <author>\n      <name>Luciano V. Del Priore</name>\n    </author>\n    <author>\n      <name>Qingyu Chen</name>\n    </author>\n  </entry>"
}