Text-to-Image Retrieval on Educontent zero-shot
14.6R@1HOMIE
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| HOMIEModel Category=MLLM-Based Models, Backbone=Qwen3-VL-8B, Evaluation Protocol=Zero-shot2025.02 | 14.6 | 31.5 | 40.5 | |
| HOMIEModel Category=MLLM-Based Models, Backbone=Qwen3-VL-2B, Evaluation Protocol=Zero-shot2025.02 | 12.7 | 29.7 | 38.3 | |
| MUSKModel Category=Pathology Dual-Encoder Models, Evaluation Protocol=Zero-shot2025.02 | 9 | 28.2 | 38.2 |