Look Where It Matters: High-Resolution Crops Retrieval for Efficient VLMs

Benchmark Model Rank Results
visual-question-answering-on-v-benchAwaRes#21Accuracy: 71.2
visual-question-answering-vqa-on-realworldqaAwaRes#56Accuracy: 68.5
chart-question-answering-on-chartqaAwaRes#671:1 Accuracy: 80.64