MetaphorStar: Image Metaphor Understanding and Reasoning with End-to-End Visual Reinforcement Learning

Benchmark Model Rank Results
multimodal-reasoning-on-visulogicMetaphorStar-7B#6Accuracy: 26.9
visual-question-answering-vqa-on-seed-seedbenchMetaphorStar-32B#6Accuracy: 77.6
multimodal-reasoning-on-mathvisionMetaphorStar-32B#42Accuracy: 38.1
vision-language-understanding-on-mmstarMetaphorStar-32B#46Accuracy: 68.5