List your company

Object Hallucination Evaluation on POPE (Accuracy, F1, Precision, Recall)

90.81AccuracyVISE
87.461288.330689.290.0694Jun 25, 2026

Evaluation results

MethodLinks
VISEModel=Qwen3-VL-32B-Instruct2026.0690.8189.9293.4586.65
VisionZero-CLEVRModel=Qwen3-VL-32B-Instruct2026.0690.7389.8693.3986.59
VisionZero-RWModel=Qwen3-VL-32B-Instruct2026.0690.6689.893.3186.54
EvoLMMModel=Qwen3-VL-32B-Instruct2026.0690.5989.7293.2386.46
VisionZero-ChartModel=Qwen3-VL-32B-Instruct2026.0690.5189.6493.1486.39
VisPlayModel=Qwen3-VL-32B-Instruct2026.0690.4589.5793.0786.32
iReasonerModel=Qwen3-VL-32B-Instruct2026.0690.3989.4992.9586.28
BaseModel=Qwen3-VL-32B-Instruct2026.0690.3589.4592.9686.2
VISEModel=Qwen3-VL-8B-Instruct2026.0690.3289.3292.7386.15
VisPlayModel=Qwen3-VL-8B-Instruct2026.0690.2389.1592.7385.84
VisionZero-CLEVRModel=Qwen3-VL-8B-Instruct2026.0690.1889.0692.7285.68
EvoLMMModel=Qwen3-VL-8B-Instruct2026.0690.1388.9892.7385.52
iReasonerModel=Qwen3-VL-8B-Instruct2026.0690.0788.8992.7285.36
VISEModel=Qwen3-VL-2B-Instruct2026.0690.0389.2293.1385.63
VisionZero-RWModel=Qwen3-VL-8B-Instruct2026.0690.0288.892.7285.2
VisionZero-ChartModel=Qwen3-VL-8B-Instruct2026.0689.9588.6992.785.01
BaseModel=Qwen3-VL-8B-Instruct2026.0689.9188.6192.7184.86
VISEModel=Qwen3-VL-4B-Instruct2026.0689.8688.9892.4785.74
VisionZero-CLEVRModel=Qwen3-VL-4B-Instruct2026.0689.8288.0293.7182.98
BaseModel=Qwen3-VL-4B-Instruct2026.0689.7388.4592.5784.68
VisionZero-RWModel=Qwen3-VL-4B-Instruct2026.0689.6789.2593.885.12
VisionZero-CLEVRModel=Qwen3-VL-2B-Instruct2026.0689.6589.0193.0985.27
iReasonerModel=Qwen3-VL-4B-Instruct2026.0689.6488.5993.1584.46
EvoLMMModel=Qwen3-VL-4B-Instruct2026.0689.5388.5693.0784.39
VisPlayModel=Qwen3-VL-4B-Instruct2026.0689.4288.9292.8984.99
VisionZero-ChartModel=Qwen3-VL-2B-Instruct2026.0689.3588.7591.8685.84
VisPlayModel=Qwen3-VL-2B-Instruct2026.0689.3288.7292.8684.93
BaseModel=Qwen3-VL-2B-Instruct2026.0689.0188.3792.4584.63
VisionZero-ChartModel=Qwen3-VL-4B-Instruct2026.0688.7588.7792.7984.93
VisionZero-RWModel=Qwen3-VL-2B-Instruct2026.0688.787.8493.0583.18
iReasonerModel=Qwen3-VL-2B-Instruct2026.0687.788.5692.1285.26
EvoLMMModel=Qwen3-VL-2B-Instruct2026.0687.5988.5192.0585.17