Robust-R1: Degradation-Aware Reasoning for Robust Visual Understanding
Open paper
Benchmark
Model
Rank
Results
visual-question-answering-vqa-on-realworldqa
Robust-R1 (SFT and RL)
#61
Accuracy: 67.71
vision-language-understanding-on-mmstar
Robust-R1 (SFT and RL)
#86
Accuracy: 56.86
Rank counts only results with a code link.