Hallucination Evaluation on EventHallusion binary QA (test)
0.655AccuracySmartSight
Evaluation Results
| Method | Links | |
|---|---|---|
| SmartSightBase Model=Video-R12025.12 | 0.655 | |
| VCDBase Model=Video-R12025.12 | 0.654 | |
| TCDBase Model=Video-R12025.12 | 0.649 | |
| SmartSightBase Model=Qwen2.5-VL-7B2025.12 | 0.631 | |
| Video-R1Decoding Method=Greedy2025.12 | 0.626 | |
| DINO-HEALBase Model=Video-R12025.12 | 0.626 | |
| VCDBase Model=Qwen2.5-VL-7B2025.12 | 0.619 | |
| TCDBase Model=Qwen2.5-VL-7B2025.12 | 0.614 | |
| Qwen2.5-VL-7BDecoding Method=Greedy2025.12 | 0.604 | |
| DINO-HEALBase Model=Qwen2.5-VL-7B2025.12 | 0.568 | |
| SmartSightBase Model=LLaVA-NEXT-Video-7B2025.12 | 0.558 | |
| TCDBase Model=LLaVA-NEXT-Video-7B2025.12 | 0.557 | |
| DINO-HEALBase Model=LLaVA-NEXT-Video-7B2025.12 | 0.539 | |
| LLaVA-NEXT-Video-7BDecoding Method=Greedy2025.12 | 0.534 | |
| VCDBase Model=LLaVA-NEXT-Video-7B2025.12 | 0.531 |