List your company

Object Hallucination Evaluation on GQA (Random)

90.13AccuracyCAI
74.11478.27282.4386.588Dec 22, 2025Jan 22, 2026Feb 23, 2026Mar 26, 2026Apr 27, 2026May 28, 2026Jun 29, 2026

Evaluation results

MethodLinks
CAIBackbone=Qwen-VL2026.0690.1389.87
MESAModel=LLaVA-v1.52026.0489.589.45
HDDModel=LLaVA-NeXT2025.1289.389.4
ICTModel=LLaVA-v1.52026.0489.2688.78
CAIBackbone=LLaVA-1.52026.0689.189.15
VAFModel=LLaVA-v1.52026.048988.63
RITUALModel=LLaVA-NeXT2025.1288.788.9
VTIModel=LLaVA-v1.52026.0488.3788.02
VCDBackbone=Qwen-VL2026.0688.3388.26
ONLYBackbone=Qwen-VL2026.0688.1387.54
CAIBackbone=InstructBLIP2026.0688.187.84
MESAModel=Qwen-VL2026.0487.9786.67
NulluModel=LLaVA-v1.52026.0487.788
HDDModel=LLaVA-1.52025.1287.287.6
RITUALModel=LLaVA-1.52025.1286.887.2
VCDModel=LLaVA-v1.52026.0486.6586.99
ICDModel=Qwen-VL2026.0486.284.62
ONLYBackbone=LLaVA-1.52026.0686.287.2
VAFModel=Qwen-VL2026.0486.1186.11
VCDModel=LLaVA-1.52025.1286.187.1
ICDModel=LLaVA-v1.52026.0485.985.29
PAIBackbone=Qwen-VL2026.0685.984.77
VCDModel=LLaVA-NeXT2025.1285.785.9
VCDModel=Qwen-VL2026.0485.5985.33
ICTModel=Qwen-VL2026.0485.3386.12
VanillaModel=LLaVA-v1.52026.0484.9686.25
VTIModel=Qwen-VL2026.0484.3381.94
HDDModel=InstructBLIP2025.1284.184.9
ONLYBackbone=InstructBLIP2026.0683.8785.02
RegularBackbone=Qwen-VL2026.0683.8382.6
SamplingModel=LLaVA-1.52025.1283.582.7
RITUALModel=InstructBLIP2025.1283.584.3
PAIBackbone=LLaVA-1.52026.0683.3785.03
VCDBackbone=InstructBLIP2026.0683.3783.78
SamplingModel=LLaVA-NeXT2025.1283.181.2
VCDModel=InstructBLIP2025.1282.883.2
RegularBackbone=LLaVA-1.52026.0682.0383.86
VCDBackbone=LLaVA-1.52026.0681.783.99
PAIBackbone=InstructBLIP2026.0681.2382.67
RegularBackbone=InstructBLIP2026.0680.1781.55
SamplingModel=InstructBLIP2025.1279.580.4
VanillaModel=Qwen-VL2026.0474.7666.6
NulluModel=Qwen-VL2026.0474.7365.55