AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

VisionUnite: A Vision-Language Foundation Model for Ophthalmology Enhanced With Clinical Knowledge

Zihan Li, Diping Song, Zefeng Yang, Deming Wang, Fei Li, Xiulan Zhang, Paul E. Kinahan, Yu Qiao

TPAMI 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Multimodal LLM benchmarks · 33 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Vision Projector to CLIP loss plus Vision Adapter (ablation)VisionUnite · Fundus multi-round VQA (SBERT) component ablationSBERT similarity (%): 79.06 -> 79.94Improved
add Vision Adapter with classification loss to CLIP plus Projector (ablation)VisionUnite · Fundus multi-round VQA (SBERT) component ablationSBERT similarity (%): 77.35 -> 79.94Improved
add CLIP contrastive loss to Adapter plus Projector (ablation)VisionUnite · Fundus multi-round VQA (SBERT) component ablationSBERT similarity (%): 78.30 -> 79.94Improved
add Vision Adapter with classification loss to CLIP loss alone (ablation)VisionUnite · Fundus multi-round VQA (SBERT) component ablationSBERT similarity (%): 75.14 -> 79.06Improved
add MS COCO pretraining (ablation)VisionUnite · Fundus multi-round VQA (SBERT) pretraining ablationSBERT similarity (%): 74.69 -> 77.65Improved
add PMC-OA pretraining (ablation)VisionUnite · Fundus multi-round VQA (SBERT) pretraining ablationSBERT similarity (%): 74.69 -> 76.76Improved
add combined pretraining datasets (ablation)VisionUnite · Fundus multi-round VQA (SBERT) pretraining ablationSBERT similarity (%): 74.69 -> 79.38Improved
VisionUnite versus junior ophthalmologistVisionUnite · Fundus multi-round VQA (180 images, expert-rated)diagnostic accuracy: .735 -> .780No clear effect
VisionUnite versus GPT-4VVisionUnite · Fundus multi-round VQA (180 images, expert-rated)diagnostic accuracy: .498 -> .780Improved
fine-tuned VisionUnite versus fine-tuned LLaVAVisionUnite · Fundus multi-round VQA (SBERT similarity)SBERT similarity (%): 74.26 -> 79.94Improved
VisionUnite versus best open-source baseline on multiple-choiceVisionUnite · Fundus multiple-choice (25 conditions)accuracy (%): 85.22 -> 94.36Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy