# FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs

arXiv: https://arxiv.org/abs/2609.33158 · HTML: https://mlsanity.com/paper/2609.33158/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 21 of its 40 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [Med-Flamingo: a Multimodal Medical Few-shot Learner](https://mlsanity.com/paper/2307.15189/) (2023)
- [REFUGE Challenge: A Unified Framework for Evaluating Automated Methods for Glaucoma Assessment from Fundus Photographs](https://mlsanity.com/paper/1910.03667/) (2019)
- [A Foundation Language-Image Model of the Retina (FLAIR): Encoding Expert Knowledge in Text Supervision](https://mlsanity.com/paper/2308.07898/) (2023)
- [MedGemma 1.5 Technical Report](https://mlsanity.com/paper/2604.05081/) (2026)
- [Grader variability and the importance of reference standards for evaluating machine learning models for diabetic retinopathy](https://mlsanity.com/paper/1710.01711/) (2017)
- [G1020: A Benchmark Retinal Fundus Image Dataset for Computer-Aided Glaucoma Detection](https://mlsanity.com/paper/2006.09158/) (2020)
- [RET-CLIP: A Retinal Image Foundation Model Pre-trained with Clinical Diagnostic Reports](https://mlsanity.com/paper/2405.14137/) (2024)
- [VisionFM: a Multi-Modal Multi-Task Vision Foundation Model for Generalist Ophthalmic Artificial Intelligence](https://mlsanity.com/paper/2310.04992/) (2023)
- [MultiEYE: Dataset and Benchmark for OCT-Enhanced Retinal Disease Recognition from Fundus Images](https://mlsanity.com/paper/2412.09402/) (2024)
- [LMOD: A Large Multimodal Ophthalmology Dataset and Benchmark for Large Vision-Language Models](https://mlsanity.com/paper/2410.01620/) (2024)
- [MIRAGE: Multimodal foundation model and benchmark for comprehensive retinal OCT image analysis](https://mlsanity.com/paper/2506.08900/) (2025)
- [FunBench: Benchmarking Fundus Reading Skills of MLLMs](https://mlsanity.com/paper/2503.00901/) (2025)
- [Fair Foundation Models for Medical Image Analysis: Challenges and Perspectives](https://mlsanity.com/paper/2502.16841/) (2025)

## Abstract (arXiv)

Progress in AI-based retinal image analysis has advanced with foundation models, yet evaluating their reliability remains challenging. Performance reported on a single dataset does not capture how models behave under dataset shift, across clinical definitions, or for different patient subgroups. This limitation is particularly critical in medical imaging analysis, where robustness, calibration, and fairness are essential for safe deployment. We introduce FOCUS (Foundation Ophthalmic Cross-Dataset Understanding under Shift), a cross-dataset benchmark for evaluating retinal fundus models that considers vision-only encoder models (VM), vision-language dual-encoder models (VLM), and multimodal large language models (MLLM). FOCUS harmonizes binary diabetic retinopathy, referable diabetic retinopathy, and glaucomatous optic neuropathy tasks across ten public datasets spanning diverse geographies, acquisition conditions, and label protocols. The benchmark evaluates models through a unified analysis layer that measures ranking performance, calibration, subgroup disparities, and image-quality robustness. We present a large-scale evaluation covering 532 base configurations and 228 MLLM configurations adapted through supervised fine-tuning with low-rank adaptation (LoRA). Results show that no model family consistently dominates across tasks and datasets: general VM encoders achieve the strongest average ranking performance, medical MLLMs are competitive but variable, and dual encoder VLMs benefit substantially from lightweight adaptation. Fine-tuning improves in-domain performance but exhibits heterogeneous transfer to external datasets, particularly in calibration. These findings demonstrate that retinal model evaluation is inherently multidimensional. FOCUS provides a practical framework and public benchmark to assess generalization, reliability, and robustness beyond single-dataset leaderboards