How Should Multimodal Information Be Chunked for Complex Question Answering?

Kangheng Liang

Browse the full SIGIR paper archive.