List your company

Multimodal Hallucination Evaluation on MME

720Total ScoreCAI
419.7832497.7241575.665653.6059Apr 27, 2026May 7, 2026May 18, 2026May 28, 2026Jun 8, 2026Jun 18, 2026Jun 29, 2026

Evaluation results

MethodLinks
CAI2026.06720————
CAIGLA-gated2026.06700————
Active-LookModel=Qwen3-VL-8B2026.04698.33195170148.33185
PAI2026.06691.66————
Qwen-3.52026.06685————
BaselineModel=Qwen3-VL-8B2026.04673.33190165143.33175
ILVADBase Model=LLaVA-1.5-7B2026.05641.66195153.33133.33160
VARBase Model=LLaVA-1.5-7B2026.05636.66185143.33143.33165
GreedyBase Model=LLaVA-1.5-7B2026.05631.66195158.33123.33155
SPARCBase Model=LLaVA-1.5-7B2026.05629.99195158.33123.33153.33
VHRBase Model=LLaVA-1.5-7B2026.05629.99195158.33123.33153.33
Active-LookModel=InternVL2-8B2026.04626.66185130131.66180
ONLYBase Model=LLaVA-1.5-7B2026.05626.66190150121.66165
VAFBase Model=LLaVA-1.5-7B2026.05624.99185151.66133.33155
AGLABase Model=LLaVA-1.5-7B2026.05621.66190153.33118.33160
VCDBase Model=LLaVA-1.5-7B2026.05614.99190163.33118.33143.33
CODEBase Model=LLaVA-1.5-7B2026.05609.99195131.66133.33150
BeamBase Model=LLaVA-1.5-7B2026.05599.99195136.66123.33145
BaselineModel=InternVL2-8B2026.04568.33180128.33120140
Active-LookModel=LLaVA-1.5-7B2026.04516.66168.33121.66116.66110
OPERAModel=LLaVA-1.5-7B2026.04497.33168.33110113.66105.33
VAFModel=LLaVA-1.5-7B2026.04485158.33111.33113.66101.66
VCDModel=LLaVA-1.5-7B2026.0444615098.3393.33104.33
BaselineModel=LLaVA-1.5-7B2026.04431.33158.3391.339091.66