Text-rich VQA

The study investigates the advantages and bottlenecks of Large Language Models (LLMs) in addressing Text-rich Visual Question Answering (VQA) tasks. These tasks involve both image comprehension and text recognition. The…