Action Reasoning on Video-MME Long (test)
46.67AccuracyEC-RAG
Evaluation Results
| Method | Links | |
|---|---|---|
| EC-RAGBackbone=Qwen3-VL-8B2026.05 | 46.67 | |
| Qwen3-VL-8BBackbone=Qwen3-VL-8B, Configuration=Baseline2026.05 | 43.89 |
Loading the SOTA2 catalog…
| Method | Links | |
|---|---|---|
| EC-RAGBackbone=Qwen3-VL-8B2026.05 | 46.67 | |
| Qwen3-VL-8BBackbone=Qwen3-VL-8B, Configuration=Baseline2026.05 | 43.89 |