Referring Object Classification on LVIS In-Domain
73.39AccuracyLLaVA-1.5 + Blur
Evaluation Results
| Method | Links | |
|---|---|---|
| LLaVA-1.5 + BlurInput format=Box, Training strategy=Training-Free Methods2026.02 | 73.39 | |
| LLaVA-1.5 + ControlMLLM++Input format=Mask, Training strategy=Training-Free Methods2026.02 | 73 | |
| Ferret-7BInput format=Mask, Training strategy=Training Methods2026.02 | 72.39 | |
| Ferret-7BInput format=Box, Training strategy=Training Methods2026.02 | 71.71 | |
| Ferret-7BInput format=Scribble, Training strategy=Training Methods2026.02 | 71.58 | |
| LLaVA-1.5 + BlurInput format=Mask, Training strategy=Training-Free Methods2026.02 | 71.32 | |
| LLaVA-1.5 + ControlMLLM++Input format=Box, Training strategy=Training-Free Methods2026.02 | 71.19 | |
| Ferret-7BInput format=Point, Training strategy=Training Methods2026.02 | 68.54 | |
| LLaVA-1.5 + ControlMLLM++Input format=Point, Training strategy=Training-Free Methods2026.02 | 65.13 | |
| LLaVA-1.5 + ControlMLLM++Input format=Scribble, Training strategy=Training-Free Methods2026.02 | 65.12 | |
| Shikra-7BInput format=Box, Training strategy=Training Methods2026.02 | 64.6 | |
| LLaVA-1.5 + ControlMLLMInput format=Mask, Training strategy=Training-Free Methods2026.02 | 60.79 | |
| LLaVA-1.5 + ControlMLLMInput format=Box, Training strategy=Training-Free Methods2026.02 | 60.59 | |
| LLaVA-1.5 + ControlMLLMInput format=Point, Training strategy=Training-Free Methods2026.02 | 58.85 | |
| GPT4-ROIInput format=Box, Training strategy=Training Methods2026.02 | 58.59 | |
| LLaVA-1.5 + ControlMLLMInput format=Scribble, Training strategy=Training-Free Methods2026.02 | 58.33 | |
| LLaVA-1.5 + ColorInput format=Mask, Training strategy=Training-Free Methods2026.02 | 56.72 | |
| Shikra-7BInput format=Point, Training strategy=Training Methods2026.02 | 56.27 | |
| Kosmos-2Input format=Box, Training strategy=Training Methods2026.02 | 55.17 | |
| LLaVA-1.5 + ColorInput format=Box, Training strategy=Training-Free Methods2026.02 | 55.1 | |
| LLaVA-1.5Input format=Box, Training strategy=Training-Free Methods2026.02 | 54.72 | |
| LLaVA-1.5Input format=Mask, Training strategy=Training-Free Methods2026.02 | 54.72 | |
| LLaVA-1.5Input format=Scribble, Training strategy=Training-Free Methods2026.02 | 54.72 | |
| LLaVA-1.5Input format=Point, Training strategy=Training-Free Methods2026.02 | 54.72 | |
| LLaVA-1.5 + Edit AttInput format=Mask, Training strategy=Training-Free Methods2026.02 | 37.08 | |
| LLaVA-1.5 + Edit AttInput format=Box, Training strategy=Training-Free Methods2026.02 | 36.24 |