List your company
ResearchFormerly WizWand

Image Captioning on Infrared-COCO (test)

72.87Semantic ConsistencyClean
42.60650.46358.3266.177Apr 3, 2026
Updated 6mo ago

Evaluation results

MethodLinks
CleanImage Encoder=EVA-CLIP ViT-G/14, Models=InstructBLIP FlanT5XL (4.1B)2026.0472.87
CleanImage Encoder=EVA-CLIP ViT-G/14, Models=Blip-2 FlanT5XL (4.1B)2026.0470.17
CleanImage Encoder=OpenAI CLIP ViT-L/14, Models=Blip-2 FlanT5XL ViT-L (3.4B)2026.0468.87
CleanImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.6 (13B)2026.0468.01
QR codeImage Encoder=EVA-CLIP ViT-G/14, Models=InstructBLIP FlanT5XL (4.1B)2026.0466.77
CleanImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.5 (7B)2026.0466.03
QR codeImage Encoder=EVA-CLIP ViT-G/14, Models=Blip-2 FlanT5XL (4.1B)2026.0464.37
QR codeImage Encoder=OpenAI CLIP ViT-L/14, Models=Blip-2 FlanT5XL ViT-L (3.4B)2026.0464.03
HCBImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.6 (13B)2026.0463.94
AdvGridImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.6 (13B)2026.0463.65
CleanImage Encoder=OpenAI CLIP ViT-L/14, Models=OpenFlamingo (3B)2026.0463.17
HCBImage Encoder=EVA-CLIP ViT-G/14, Models=Blip-2 FlanT5XL (4.1B)2026.0462.45
HCBImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.5 (7B)2026.0461.99
QR codeImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.5 (7B)2026.0461.59
QR codeImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.6 (13B)2026.0461.09
AdvGridImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.5 (7B)2026.0461
AdvGridImage Encoder=OpenAI CLIP ViT-L/14, Models=Blip-2 FlanT5XL ViT-L (3.4B)2026.0460.96
HCBImage Encoder=EVA-CLIP ViT-G/14, Models=InstructBLIP FlanT5XL (4.1B)2026.0460.63
AdvGridImage Encoder=EVA-CLIP ViT-G/14, Models=Blip-2 FlanT5XL (4.1B)2026.0459.93
AdvGridImage Encoder=EVA-CLIP ViT-G/14, Models=InstructBLIP FlanT5XL (4.1B)2026.0459.13
HCBImage Encoder=OpenAI CLIP ViT-L/14, Models=Blip-2 FlanT5XL ViT-L (3.4B)2026.0458.91
QR codeImage Encoder=OpenAI CLIP ViT-L/14, Models=OpenFlamingo (3B)2026.0458.37
AdvGridImage Encoder=OpenAI CLIP ViT-L/14, Models=OpenFlamingo (3B)2026.0454.93
UCGPImage Encoder=EVA-CLIP ViT-G/14, Models=InstructBLIP FlanT5XL (4.1B)2026.0453.35
UCGPImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.6 (13B)2026.0452.77
UCGPImage Encoder=EVA-CLIP ViT-G/14, Models=Blip-2 FlanT5XL (4.1B)2026.0451.77
HCBImage Encoder=OpenAI CLIP ViT-L/14, Models=OpenFlamingo (3B)2026.0451.37
UCGPImage Encoder=OpenAI CLIP ViT-L/14, Models=Blip-2 FlanT5XL ViT-L (3.4B)2026.0451.2
UCGPImage Encoder=OpenAI CLIP ViT-L/14, Models=LLaVA-1.5 (7B)2026.0448.79
UCGPImage Encoder=OpenAI CLIP ViT-L/14, Models=OpenFlamingo (3B)2026.0443.77