MDocRAG-RL: Empowering Multi-Modal Document RAG via Complex Visual Reasoning with Reinforcement Learning.

Zhongyu Wang

Browse the full ACL paper archive.