Plan-Grounded Answer Generation on InstructionVidDial (test)
75.3ROUGE-LVIGiA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VIGiA2026.02 | 75.3 | 76.67 | 88.72 | |
| MM-PlanLLM2026.02 | 58.85 | 59.34 | 80.03 | |
| Qwen 3 VL2026.02 | 44.71 | 46.31 | 71.94 | |
| LLaVA-1.52026.02 | 42.47 | 40.84 | 70.55 | |
| InternVL 3.52026.02 | 41.62 | 46.2 | 69.22 | |
| LLaVA-OV2026.02 | 41.18 | 40.85 | 70.7 | |
| Idefics22026.02 | 37.16 | 45.09 | 67.55 | |
| Qwen 2.5 VL2026.02 | 31.61 | 41.51 | 63.91 |