Video Understanding on Multiple Aggregate
69.8Average ScoreReMoRa
Evaluation results
| Method | Links | |
|---|---|---|
| ReMoRaLLM backbone=Qwen2-7B2026.02 | 69.8 | |
| BIMBALLM backbone=Qwen2-7B2026.02 | 68.9 | |
| LLaVA-VideoLLM backbone=Qwen2-7B2026.02 | 68.7 | |
| Qwen2.5-VLLLM backbone=Qwen2.5-7B2026.02 | 67.4 | |
| LLaVA-OneVisionLLM backbone=Qwen2-7B2026.02 | 63.2 | |
| LongVULLM backbone=Qwen2-7B2026.02 | 63 | |
| KangarooLLM backbone=Llama-3-8B2026.02 | 59.9 | |
| Qwen2-VLLLM backbone=Qwen2-7B2026.02 | 59.5 | |
| Video-XLLLM backbone=Qwen2-7B2026.02 | 56.6 | |
| LongVALLM backbone=Qwen2-7B2026.02 | 55.1 | |
| EMALLM backbone=Qwen2-7B2026.02 | 52.5 | |
| PLLaVALLM backbone=Vicuna-7B2026.02 | 50 | |
| Video-LLaMA2LLM backbone=Mistral-7B2026.02 | 49.9 | |
| Video-LLaVALLM backbone=Vicuna-7B2026.02 | 48.7 | |
| Video-LaVITLLM backbone=LaVIT-7B2026.02 | 47.9 | |
| VideoChat2LLM backbone=Vicuna-7B2026.02 | 46.5 | |
| LLaVA-NeXT-VideoLLM backbone=Qwen1.5-7B2026.02 | 46.3 | |
| LLaMA-VIDLLM backbone=Vicuna-7B2026.02 | 38.9 |