Video Captioning on PE-Video (RCC)
7.67RelevanceRank
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| RankBackbone=PerceptionLM2026.03 | 7.67 | 7.72 | 7.93 | 9.49 | |
| DPOBackbone=PerceptionLM2026.03 | 7.57 | 7.59 | 7.78 | 9.43 | |
| MPOBackbone=PerceptionLM2026.03 | 7.57 | 7.59 | 7.8 | 9.42 | |
| RankBackbone=Qwen2.5-VL2026.03 | 7.46 | 7.4 | 7.75 | 9.4 | |
| MPOBackbone=Qwen2.5-VL2026.03 | 7.35 | 7.27 | 7.66 | 9.34 | |
| DPOBackbone=Qwen2.5-VL2026.03 | 7.33 | 7.25 | 7.6 | 9.36 | |
| SFTBackbone=Qwen2.5-VL2026.03 | 7.31 | 7.24 | 7.64 | 9.35 | |
| PretrainedBackbone=Qwen2.5-VL2026.03 | 6.74 | 6.12 | 6.63 | 8.69 | |
| PretrainedBackbone=PerceptionLM2026.03 | 6.72 | 6.04 | 5.82 | 6.98 | |
| SFTBackbone=PerceptionLM2026.03 | 6.35 | 5.6 | 6.07 | 8.15 |