List your company

Hallucination Evaluation on MS-COCO

91.6Accuracy (Rand.)RITUAL
85.98487.44288.990.358Mar 17, 2026

Evaluation results

MethodLinks
RITUALBackbone=InternVL3.52026.0391.689.0387.48
KestrelBackbone=Qwen3-VL2026.0391.5389.387.53
OPERABackbone=InternVL3.52026.0391.5389.1887.55
DeGFBackbone=InternVL3.52026.0391.4389.1287.37
VCDBackbone=InternVL3.52026.0391.3589.2287.6
KestrelBackbone=InternVL3.52026.0391.2789.2788.1
WoodpeckerBackbone=InternVL3.52026.0391.289.1187.5
Qwen3-VL agentBackbone=Qwen3-VL2026.0391.0388.0686.13
BaselineBackbone=InternVL3.52026.0390.7788.185.73
OPERABackbone=Qwen3-VL2026.0390.588.8387.5
VCDBackbone=Qwen3-VL2026.0390.488.887.41
DeGFBackbone=Qwen3-VL2026.0390.3388.1686.9
WoodpeckerBackbone=Qwen3-VL2026.0389.9788.0387.1
BaselineBackbone=Qwen3-VL2026.038986.9286.2
RITUALBackbone=Qwen3-VL2026.0386.283.6782.27