MMLongBench-Doc

MMLongBench-Doc is a benchmark for reading and parsing real documents. The official description says: "MMLongBench-Doc - long document understanding benchmark evaluating comprehension and retrieval over extended documents". Below we describe in simple words what benchmarks of this type generally examine.

MMLongBench-Doc measures reading and parsing real documents. In detail, it checks:

• extracting structured text from scans, PDFs and photos
• keeping tables, headings and reading order intact
• answering comprehension questions about the document content
• handling noisy layouts, handwriting and multi-page documents

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: MMLongBench-Doc - long document understanding benchmark evaluating comprehension and retrieval over extended documents.

What is the MMLongBench-Doc benchmark?

MMLongBench-Doc is a benchmark for reading and parsing real documents. The official description says: 'MMLongBench-Doc - long document understanding benchmark evaluating comprehension and retrieval over extended documents'. Below we describe in simple words what benchmarks of this type generally examine.

What does MMLongBench-Doc test?

MMLongBench-Doc mainly measures reading and parsing real documents. Concretely, it checks: extracting structured text from scans, PDFs and photos; keeping tables, headings and reading order intact; answering comprehension questions about the document content.

How is the MMLongBench-Doc score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Where do I get MMLongBench-Doc data and scores?

The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/mmlongbench-doc/scores) or pull it directly from the Open LLM Radar API with any HTTP client.