List your company

Hallucination Evaluation on GQA

91.8Accuracy (Random)OPERA
86.662487.996289.3390.6638Mar 17, 2026

Evaluation results

MethodLinks
OPERABackbone=Qwen3-VL2026.0391.887.1183.3
KestrelBackbone=Qwen3-VL2026.0391.6790.3386.27
VCDBackbone=InternVL3.52026.0391.685.0783.37
KestrelBackbone=InternVL3.52026.0391.5789.8786.53
VCDBackbone=Qwen3-VL2026.0391.5685.7681.93
DeGFBackbone=InternVL3.52026.0391.4885.0683.2
OPERABackbone=InternVL3.52026.0391.4585.283.31
WoodpeckerBackbone=InternVL3.52026.0391.4385.1683.26
Qwen3-VL agentBackbone=Qwen3-VL2026.0391.4181.8178.3
RITUALBackbone=InternVL3.52026.0391.3985.1883.29
WoodpeckerBackbone=Qwen3-VL2026.0391.2786.2782.77
DeGFBackbone=Qwen3-VL2026.0391.1383.7982
BaselineBackbone=Qwen3-VL2026.0390.783.6381.5
BaselineBackbone=InternVL3.52026.0389.7784.181.31
RITUALBackbone=Qwen3-VL2026.0386.8682.378.23