List your company

Hallucination Evaluation on MME Hallucination 9

200ExistenceQwen3-VL agent
194.8196.15197.5198.85Mar 17, 2026

Evaluation results

MethodLinks
Qwen3-VL agentBackbone=Qwen3-VL [3]2026.03200181.67168.33193.33743.33
KestrelBackbone=Qwen3-VL [3]2026.03200186.67180193.33760
BaselineBackbone=InternVL3.5 [31]2026.03200175175193.33743.33
VCDBackbone=InternVL3.5 [31]2026.03200175175193.33736.66
WoodpeckerBackbone=InternVL3.5 [31]2026.03200166.67161.67186.67715.01
KestrelBackbone=InternVL3.5 [31]2026.03200186.67181.67195763.34
BaselineBackbone=Qwen3-VL [3]2026.03195175168.33193.33731.66
VCDBackbone=Qwen3-VL [3]2026.03195180168.33193.33736.66
WoodpeckerBackbone=Qwen3-VL [3]2026.03195173.33168.33195731.66
RITUALBackbone=Qwen3-VL [3]2026.03195180168.33193.33736.66
OPERABackbone=Qwen3-VL [3]2026.03195180168.33200743.33
DeGFBackbone=Qwen3-VL [3]2026.03195181.67166.67188.33732.67
RITUALBackbone=InternVL3.5 [31]2026.03195175175193.33738.33
OPERABackbone=InternVL3.5 [31]2026.03195173.33175195738.33
DeGFBackbone=InternVL3.5 [31]2026.03195175168.33188.33726.66