MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizations

TL;DR

This paper introduces MMLongBench-Doc, a comprehensive benchmark dataset for evaluating large vision-language models on long, multi-modal documents, revealing current models' significant challenges in understanding lengthy, complex documents.

Contribution

It provides the first long-context, multi-modal benchmark with expert annotations for document understanding, highlighting the limitations of existing LVLMs in handling long, multi-source, multi-page documents.

Findings

Current LVLMs perform poorly on long-context document understanding.

GPT-4o achieves only 42.7% F1 score on the benchmark.

Most models underperform compared to LLMs with OCR data.

Abstract

Understanding documents with rich layouts and multi-modal components is a long-standing and practical task. Recent Large Vision-Language Models (LVLMs) have made remarkable strides in various tasks, particularly in single-page document understanding (DU). However, their abilities on long-context DU remain an open problem. This work presents MMLongBench-Doc, a long-context, multi-modal benchmark comprising 1,062 expert-annotated questions. Distinct from previous datasets, it is constructed upon 130 lengthy PDF-formatted documents with an average of 49.4 pages and 20,971 textual tokens. Towards comprehensive evaluation, answers to these questions rely on pieces of evidence from (1) different sources (text, image, chart, table, and layout structure) and (2) various locations (i.e. page number). Moreover, 33.2% of the questions are cross-page questions requiring evidence across multiple pages. 22.8% of the questions are designed to be unanswerable for detecting potential hallucinations. Experiments on 14 LVLMs demonstrate that long-context DU greatly challenges current models. Notably, the best-performing model, GPT-4o, achieves an F1 score of only 42.7%, while the second-best, GPT-4V, scores 31.4%. Furthermore, 12 LVLMs (all except GPT-4o and GPT-4V) even present worse performance than their LLM counterparts which are fed with lossy-parsed OCR documents. These results validate the necessity of future research toward more capable long-context LVLMs. Project Page: https://mayubo2333.github.io/MMLongBench-Doc