一作SIGIR 2026Oral· Information Retrieval (Full)
Good Ranks Follow Good Answers: Unsupervised Answer-Driven Reranking for Multimodal Document QA
Keyu Zhu · Shuanghong Shen · Xianquan Wang · Kai Zhang · Shijin Wang · Qi Liu · Zhenya Huang
摘要
Multimodal Document Question Answering (MDQA) systems commonly follow a retrieve-then-answer paradigm; however, the retrieval stage often introduces substantial noise, making an effective reranking component indispensable.Existing reranker training frameworks in MDQA rely predominantly on proxy s...