Semantic Role Labeling Captioning on VidSitu
76.34CIDErCineMEC
Evaluation results
| Method | Links | |
|---|---|---|
| CineMEC2026.04 | 76.34 | |
| ClipSitu2026.04 | 61.93 | |
| LLaVA-VideoType=Instruction-tuned MLLM2026.04 | 60.1 | |
| Qwen2-VLType=Instruction-tuned MLLM2026.04 | 57.28 | |
| VILAType=Instruction-tuned MLLM2026.04 | 40.44 |