Incentivizing Vision Language Models to Search for Long Video Question Answering figure
AlphaXiv 中文概览(可滚动查看)