Spatial Understanding on SEED-Bench Spatial
66.28AccuracyCVLM
Evaluation results
| Method | Links | |
|---|---|---|
| CVLMLLM=Qwen-VL2024.02 | 66.28 | |
| CVLM (3M IKPairs, Objects=3)LLM=Vicuna-7B, Objects=32024.02 | 62.93 | |
| Qwen-VLLLM=Qwen2024.02 | 62.41 | |
| CVLM (3M IKPairs, Objects=5)LLM=Vicuna-7B, Objects=52024.02 | 60.3 | |
| InstructBLIPLLM=Flan-T5-XXL2024.02 | 60.29 | |
| CVLM (3M IKPairs, Objects=8)LLM=Vicuna-7B, Objects=82024.02 | 59.83 | |
| CVLM (3M IKPairs) w/o FKALLM=Vicuna-7B2024.02 | 59.73 | |
| CVLM (3M IKPairs, Objects=1)LLM=Vicuna-7B, Objects=12024.02 | 59.65 | |
| CVLM w/o FKALLM=Vicuna-7B2024.02 | 59.63 | |
| InstructBLIPLLM=Vicuna-7B2024.02 | 58.8 | |
| CVLMLLM=Vicuna-7B2024.02 | 58.77 | |
| InstructBLIPLLM=Flan-T5-XL2024.02 | 57.8 | |
| CVLM w/o (FKA & VKA)LLM=Vicuna-7B2024.02 | 55.11 | |
| LLaVA-v1.5†LLM=Vicuna-7B2024.02 | 55.05 | |
| MiniGPT4LLM=Vicuna-7B2024.02 | 47.4 |