Hallucination Evaluation on MME Hallucination 9
200ExistenceQwen3-VL agent
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Qwen3-VL agentBackbone=Qwen3-VL [3]2026.03 | 200 | 181.67 | 168.33 | 193.33 | 743.33 | |
| KestrelBackbone=Qwen3-VL [3]2026.03 | 200 | 186.67 | 180 | 193.33 | 760 | |
| BaselineBackbone=InternVL3.5 [31]2026.03 | 200 | 175 | 175 | 193.33 | 743.33 | |
| VCDBackbone=InternVL3.5 [31]2026.03 | 200 | 175 | 175 | 193.33 | 736.66 | |
| WoodpeckerBackbone=InternVL3.5 [31]2026.03 | 200 | 166.67 | 161.67 | 186.67 | 715.01 | |
| KestrelBackbone=InternVL3.5 [31]2026.03 | 200 | 186.67 | 181.67 | 195 | 763.34 | |
| BaselineBackbone=Qwen3-VL [3]2026.03 | 195 | 175 | 168.33 | 193.33 | 731.66 | |
| VCDBackbone=Qwen3-VL [3]2026.03 | 195 | 180 | 168.33 | 193.33 | 736.66 | |
| WoodpeckerBackbone=Qwen3-VL [3]2026.03 | 195 | 173.33 | 168.33 | 195 | 731.66 | |
| RITUALBackbone=Qwen3-VL [3]2026.03 | 195 | 180 | 168.33 | 193.33 | 736.66 | |
| OPERABackbone=Qwen3-VL [3]2026.03 | 195 | 180 | 168.33 | 200 | 743.33 | |
| DeGFBackbone=Qwen3-VL [3]2026.03 | 195 | 181.67 | 166.67 | 188.33 | 732.67 | |
| RITUALBackbone=InternVL3.5 [31]2026.03 | 195 | 175 | 175 | 193.33 | 738.33 | |
| OPERABackbone=InternVL3.5 [31]2026.03 | 195 | 173.33 | 175 | 195 | 738.33 | |
| DeGFBackbone=InternVL3.5 [31]2026.03 | 195 | 175 | 168.33 | 188.33 | 726.66 |