Object-Property Induction on VL-ICL Bench OP_IND
89.4AccuracyQwen2.5-VL-7B-Instruct
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-VL-7B-InstructVL-Data=-NA-, Params trained=8B, TTA=FT≤302025.06 | 89.4 | |
| Qwen2.5-VL-7B-InstructVL-Data=-NA-, Params trained=8B, TTA=ICL2025.06 | 84.2 | |
| Qwen2-VL-7B-InstructVL-Data=-NA-, Params trained=8B, TTA=ICL2025.06 | 69.6 | |
| LLaVA-OneVision-72BVL-Data=10.4M, Params trained=73.2B, TTA=ICL2025.06 | 69.1 | |
| LAQ-7B + MAPDVL-Data=1.3M, Params trained=24M, TTA=FT≤302025.06 | 47.7 | |
| LLaVA-OneVision-7BVL-Data=10.4M, Params trained=8B, TTA=FT≤302025.06 | 46.1 | |
| LLaVA-OneVision-7BVL-Data=10.4M, Params trained=8B, TTA=ICL2025.06 | 41.7 | |
| LAQ-7B + In-ContextPDVL-Data=1.3M, Params trained=24M, TTA=FT≤302025.06 | 30.9 | |
| LAQ-7B + In-ContextPDVL-Data=1.3M, Params trained=24M, TTA=ICL2025.06 | 20.6 | |
| LLaVA-Next-7BVL-Data=1.3M, Params trained=7.06B, TTA=FT≤302025.06 | 13.4 | |
| LLaVA v1.5-7BVL-Data=1.2M, Params trained=7B, TTA=FT≤302025.06 | 11.4 | |
| LAQ-7B + MAPDVL-Data=1.3M, Params trained=24M, TTA=ICL2025.06 | 9.6 | |
| LLaVA v1.5-7BVL-Data=1.2M, Params trained=7B, TTA=ICL2025.06 | 5.4 | |
| LLaVA-Next-7BVL-Data=1.3M, Params trained=7.06B, TTA=ICL2025.06 | 5.4 |