Tag
This paper presents MMLongBench-Doc-V2, a corrected and semantics-aware revision of the MMLongBench-Doc long-document QA benchmark, fixing annotation errors and replacing string matching with an LLM judge, along with a decision procedure for empty-set keys.
MARDoc is a memory-aware refinement agent framework for multimodal long document question answering, evaluated on MMLongBench-Doc and DocBench benchmarks using Qwen3-VL models, showing consistent improvements over MLLM-based, RAG-based, and agent-based baselines.