LLaVA-SP: Enhancing Visual Representation with Visual Spatial Tokens for MLLMs
Open paper
Benchmark
Model
Rank
Results
visual-instruction-following-on-llava-bench
LLaVA-SP-Pooling
#5
avg score: 68.3
visual-instruction-following-on-llava-bench
LLaVA-SP-Cropping
#6
avg score: 66.7
Rank counts only results with a code link.