Video-LLM Performance Calibration on Eval-OOD (EgoSchema & LongVideoBench)
5.8ECEVideoABC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VideoABCTarget Video-LLM=LLaVA-OV (7B), Model Size Category=Large size2026.05 | 5.8 | — | |
| VideoABCModel Size=–, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 8.7 | 226 | |
| VideoABCTarget Video-LLM=Qwen-3.5-VL (9B), Model Size Category=Large size2026.05 | 8.7 | — | |
| VideoABCTarget Video-LLM=Video-LLaVA 7B, Model Size Category=Large size2026.05 | 9.7 | — | |
| VideoABCTarget Video-LLM=LLaVA-Next 7B, Model Size Category=Large size2026.05 | 10.2 | — | |
| JudgeTarget Video-LLM=LLaVA-Next 7B, Model Size Category=Large size, Judge Model=LLaVA-OV (7B)2026.05 | 12.1 | — | |
| MLP-ATarget Video-LLM=LLaVA-Next 7B, Model Size Category=Large size2026.05 | 14.6 | — | |
| JudgeTarget Video-LLM=LLaVA-OV (7B), Model Size Category=Large size, Judge Model=LLaVA-OV (7B)2026.05 | 14.8 | — | |
| MLP-ATarget Video-LLM=LLaVA-OV (7B), Model Size Category=Large size2026.05 | 15.6 | — | |
| VideoABCTarget Video-LLM=Qwen-3.5-VL 4B, Model Size Category=Small size2026.05 | 15.6 | — | |
| JudgeModel Size=72B, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 16.4 | 1,802 | |
| JudgeModel Size=7B, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 17.1 | 656 | |
| JudgeTarget Video-LLM=Qwen-3.5-VL (9B), Model Size Category=Large size, Judge Model=LLaVA-OV (7B)2026.05 | 17.1 | — | |
| VideoABCTarget Video-LLM=LLaVA-OV 0.5B, Model Size Category=Small size2026.05 | 17.2 | — | |
| JudgeTarget Video-LLM=Video-LLaVA 7B, Model Size Category=Large size, Judge Model=LLaVA-OV (7B)2026.05 | 17.3 | — | |
| MLP-AModel Size=–, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 18.2 | 119 | |
| JudgeModel Size=0.5B, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 18.7 | 167 | |
| JudgeModel Size=3B, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 18.7 | 427 | |
| DirectTarget Video-LLM=Video-LLaVA 7B, Model Size Category=Large size2026.05 | 18.9 | — | |
| MLP-ATarget Video-LLM=Video-LLaVA 7B, Model Size Category=Large size2026.05 | 19.2 | — | |
| DirectModel Size=–, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 19.4 | 201 | |
| DirectTarget Video-LLM=Qwen-3.5-VL (9B), Model Size Category=Large size2026.05 | 19.4 | — | |
| MLP-ATarget Video-LLM=Qwen-3.5-VL (9B), Model Size Category=Large size2026.05 | 19.7 | — | |
| MLP-ATarget Video-LLM=Qwen-3.5-VL 4B, Model Size Category=Small size2026.05 | 19.8 | — | |
| JudgeTarget Video-LLM=LLaVA-OV 0.5B, Model Size Category=Small size, Judge Model=LLaVA-OV (7B)2026.05 | 19.8 | — | |
| DirectTarget Video-LLM=LLaVA-OV (7B), Model Size Category=Large size2026.05 | 20 | — | |
| DirectTarget Video-LLM=LLaVA-Next 7B, Model Size Category=Large size2026.05 | 20.3 | — | |
| JudgeTarget Video-LLM=Qwen-3.5-VL 4B, Model Size Category=Small size, Judge Model=LLaVA-OV (7B)2026.05 | 20.8 | — | |
| MLP-2LModel Size=9B, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 21.7 | 678 | |
| MLP-2LTarget Video-LLM=Qwen-3.5-VL (9B), Model Size Category=Large size2026.05 | 21.7 | — | |
| MLP-ATarget Video-LLM=LLaVA-OV 0.5B, Model Size Category=Small size2026.05 | 23.4 | — | |
| DirectTarget Video-LLM=Qwen-3.5-VL 4B, Model Size Category=Small size2026.05 | 24.6 | — | |
| DirectTarget Video-LLM=LLaVA-OV 0.5B, Model Size Category=Small size2026.05 | 25.6 | — | |
| MLP-2LTarget Video-LLM=LLaVA-OV (7B), Model Size Category=Large size2026.05 | 28.5 | — | |
| MLP-2LTarget Video-LLM=Video-LLaVA 7B, Model Size Category=Large size2026.05 | 29.2 | — | |
| RandomModel Size=-, Target Video-LLM=Qwen-3.5-VL (9B)2026.05 | 29.3 | 10 | |
| RandomTarget Video-LLM=Qwen-3.5-VL (9B), Model Size Category=Large size2026.05 | 29.3 | — | |
| MLP-2LTarget Video-LLM=LLaVA-Next 7B, Model Size Category=Large size2026.05 | 30.8 | — | |
| MLP-2LTarget Video-LLM=Qwen-3.5-VL 4B, Model Size Category=Small size2026.05 | 32.2 | — | |
| RandomTarget Video-LLM=LLaVA-OV (7B), Model Size Category=Large size2026.05 | 32.8 | — | |
| RandomTarget Video-LLM=LLaVA-Next 7B, Model Size Category=Large size2026.05 | 32.9 | — | |
| RandomTarget Video-LLM=Video-LLaVA 7B, Model Size Category=Large size2026.05 | 34.6 | — | |
| MLP-2LTarget Video-LLM=LLaVA-OV 0.5B, Model Size Category=Small size2026.05 | 34.7 | — | |
| RandomTarget Video-LLM=Qwen-3.5-VL 4B, Model Size Category=Small size2026.05 | 34.8 | — | |
| RandomTarget Video-LLM=LLaVA-OV 0.5B, Model Size Category=Small size2026.05 | 34.8 | — |