M-LongDoc: A Benchmark For Multimodal Super-Long Document Understanding And A Retrieval-Aware Tuning Framework

The ability to understand and answer questions over documents can be useful in many business and practical applications. However, documents often contain lengthy and diverse multimodal contents such as texts, figures, and tables, which are very time-consuming for humans to read thoroughly. Hence, there is an urgent need to develop effective and automated methods to aid humans in this task. In this work, we introduce M-LongDoc, a benchmark of 851 samples, and an automated framework to evaluate the performance of large multimodal models. We further propose a retrieval-aware tuning approach for efficient and effective multimodal document reading. Compared to existing works, our benchmark consists of more recent and lengthy documents with hundreds of pages, while also requiring open-ended explanations and not just extractive answers. To our knowledge, our training framework is the first to directly address the retrieval setting for multimodal long documents. To enhance open models, we construct a training corpus in a fully automatic manner. Experiments show that our tuning approach significantly improves the correctness of model responses by 4.6%. 1 * Yew Ken and Chaoqun were students under the Joint PhD Program between Alibaba and their corresponding university. Work done while Liying, Mahani, and Lidong were at Alibaba. † Corresponding authors. 1 Our multimodal benchmark, training corpus, and source code are publicly available at https://multimodal-documents.github.io/ .