Referring Expression Comprehension on Flickr30k Entities
83.5AccuracyCoT4DET-7B
Evaluation results
| Method | Links | |
|---|---|---|
| CoT4DET-7BType=MLLM2025.12 | 83.5 | |
| Qwen2.5-VL-7BType=MLLM2025.12 | 63.8 |
| Method | Links | |
|---|---|---|
| CoT4DET-7BType=MLLM2025.12 | 83.5 | |
| Qwen2.5-VL-7BType=MLLM2025.12 | 63.8 |