Image Editing on GIEBench (MCQA, CLIP)
24MCQA AccuracySEED-LLaMA
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SEED-LLaMALLM=Vicuna-7B, PT Data Size=77M2025.05 | 24 | 88.6 | |
| Slot-MLLM (7B)LLM=Vicuna-7B, PT Data Size=25M2025.05 | 23.9 | 94.2 | |
| LaVITLLM=LLaMA-7B, PT Data Size=100M2025.05 | 20.57 | 91.5 | |
| Slot-MLLM (14B)LLM=Qwen2.5-14B-Instruct, PT Data Size=25M2025.05 | 20.2 | 94.2 |