Benchmark RadarMergeMedBench
MergeMedBench evaluates model merging methods for medical LVLMs across eight imaging modalities, comprising 16 LoRA fine-tuned models. It provides evaluation datasets and released model checkpoints for benchmarking merging approaches.
- Released
- 2026-07-17
- Readiness
- Runnable
- Primary field
- Health & Life Sciences
Why it matters
Addresses the lack of systematic evaluation for model merging in medical imaging, enabling comparison of merging methods and serving as a practical baseline.
Motivation
Large vision-language models (LVLMs) can be adapted to specialized medical imaging tasks via parameter-efficient fine-tuning approaches such as low-rank adaptation (LoRA), leading to a growing ecosystem of expert models tailored to specific imaging modalities and clinical scenarios.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.