Fine-grained Visual Recognition on Pet-37 (test)
84.4EMSigLIP-FT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SigLIP-FTModel Category=CLIP-based2026.04 | 84.4 | 94.4 | |
| HyMOR-GemmaModel Category=HyMOR framework2026.04 | 84.4 | 94.4 | |
| HyMOR-QwenModel Category=HyMOR framework2026.04 | 84.1 | 94.2 | |
| Finedefics-8BModel Category=Fine-grained MLLMs2026.04 | 66.9 | 87.8 | |
| SigLIPModel Category=CLIP-based2026.04 | 57.8 | 76.8 | |
| BioCLIPModel Category=CLIP-based2026.04 | 54.4 | 73.5 | |
| Qwen2.5-VL-3BModel Category=General MLLMs2026.04 | 34.3 | 65.1 | |
| Gemma-3-4BModel Category=General MLLMs2026.04 | 27 | 56.8 | |
| InternVL2.5-4BModel Category=General MLLMs2026.04 | 16.4 | 51.9 |