List your company

Multimodal Perception on MME perception-related tasks

195ExistenceHIME
143156.5170183.5Aug 3, 2025Sep 5, 2025Oct 9, 2025Nov 12, 2025Dec 15, 2025Jan 18, 2026Feb 21, 2026

Evaluation results

MethodLinks
HIMEBackbone=LLaVA-1.52026.02195155.56123.33151.67130.27135.59154.5161.67120.75125—
DAMOModel=LLaVA1.5, Decoding=DAMO2025.08195150143.3165144.6134.1157.8163.81201401,513.5
NulluBackbone=LLaVA-1.52026.02190121.11105.56156.67127.55115.59147.92131.66113121.67—
VanillaModel=LLaVA1.5, Decoding=Vanilla2025.08190160138.3165140.5135156.3162.3119.31251,491.6
VCDModel=LLaVA1.5, Decoding=VCD2025.08190148.3118.3148.3131.6151.8150.3165.3125.51301,459.4
DCLAModel=LLaVA1.5, Decoding=DCLA2025.08190163.3148.3175137.4132.1156.3160.5117.31401,520.1
SPINModel=LLaVA1.5, Decoding=SPIN2025.08190163.3133.3170141.5132.9157.5165.31231251,501.9
HGAIModel=LLaVA1.5, Decoding=HGAI2025.08190160138.3165140.5135156.3160.8118.51251,489.3
MAPModel=LLaVA1.5, Decoding=MAP2025.08190165143.3170142.5138.2159.3165.3115.81401,529.3
VanillaModel=mPLUG-Owl2, Decoding=Vanilla2025.0818516578.3150163.3162.9152.8160139.8102.51,459.5
DAMOModel=mPLUG-Owl2, Decoding=DAMO2025.0818516093.3150163.3163.8153.8161.5139951,464.7
DCLAModel=mPLUG-Owl2, Decoding=DCLA2025.0818516578.3155162.2163.8153159.5139102.51,463.4
SPINModel=mPLUG-Owl2, Decoding=SPIN2025.0818516068.3160160.2166.2158.3160136.5102.51,457
HGAIModel=mPLUG-Owl2, Decoding=HGAI2025.0818516586.7150157.1162.4153.5160.8136.5102.51,459.4
MAPModel=mPLUG-Owl2, Decoding=MAP2025.0818516583.3155160.2163.5153.3162.5135.8102.51,466.4
VanillaModel=InstructBLIP, Decoding=Vanilla2025.0818558.358.3143.3152.7154.7160.8165.8112.5801,271.4
DAMOModel=InstructBLIP, Decoding=DAMO2025.0818578.368.3143.3149155.3157.3159.3116.387.51,299.5
DCLAModel=InstructBLIP, Decoding=DCLA2025.0818573.368.3143.3149.7157.6158157.3119.587.51,299.6
SPINModel=InstructBLIP, Decoding=SPIN2025.0818586.763.3143.3141.2152.9159.8160.3114.387.51,294.2
HGAIModel=InstructBLIP, Decoding=HGAI2025.0818558.356.7143.3152.7153.8159165111801,264.9
MAPModel=InstructBLIP, Decoding=MAP2025.081857063.3143.3147.3158.5159.5165.8115951,302.7
OriginalBackbone=LLaVA-1.52026.02181.67118.33104.44152.78117.23111.67144.83130.65108.9275.83—
VCDModel=mPLUG-Owl2, Decoding=VCD2025.0817514583.3150149.3162.4151147.8135.3951,394
VCDModel=InstructBLIP, Decoding=VCD2025.0814568.363.3140117127.1149.3133.810763.31,153.2