LLaVA-Llama-3

Reported on 7 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing7 results

Long-Context UnderstandingonMMNeedle
1 Image, 2*2 Stitching, Exact Accuracy· 2024-03-18
43.8
best: 94.6 (GPT-4o)
LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images arXiv:2403.11703
Long-Context UnderstandingonMMNeedle
1 Image, 4*4 Stitching, Exact Accuracy· 2024-03-18
17.5
best: 83 (GPT-4o)
LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images arXiv:2403.11703
Long-Context UnderstandingonMMNeedle
1 Image, 8*8 Stitching, Exact Accuracy· 2024-03-18
3.3
best: 29.81 (Gemini Pro 1.5)
LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images arXiv:2403.11703
Long-Context UnderstandingonMMNeedle
10 Images, 1*1 Stitching, Exact Accuracy
0
best: 97 (GPT-4o)
Long-Context UnderstandingonMMNeedle
10 Images, 2*2 Stitching, Exact Accuracy
0
best: 81.8 (GPT-4o)
Long-Context UnderstandingonMMNeedle
10 Images, 4*4 Stitching, Exact Accuracy
0
best: 26.9 (GPT-4o)
Long-Context UnderstandingonMMNeedle
10 Images, 8*8 Stitching, Exact Accuracy
0
best: 1 (GPT-4o)