pretty\name: TopoGraphRAG-Bench task\categories: - question-answering - benchmark - multimodal - document-question-answering - retrieval-augmented-generation - graph-rag - text - github.io/TopoGraphRAG-Bench/) | Code | Case Study TopoGraphRAG-Bench is a…
Dataset Card
pretty\name: TopoGraphRAG-Bench task\categories: - question-answering - benchmark - multimodal - document-question-answering - retrieval-augmented-generation - graph-rag - text - github.io/TopoGraphRAG-Bench/) | Code | Case Study TopoGraphRAG-Bench is a benchmark for layout-grounded multimodal evidence reasoning in GraphRAG. It contains 2,024 question–answer pairs over 201 long documents sampled from MMDocIR (https://huggingface.co/datasets/MMDocIR/MMDocIR\Eval\Dataset), spanning nine domains, including academic papers, legal documents, research reports, and financial reports. Each document averages approximately 45 pages and 13,000 words, with evidence grounded in text, figures, and tables…
Excerpt from the card by RrrealLL.
Details
- Repository
- diandianone123/topographrag-bench
- Publisher
- RrrealLL
- Task category
- Question answering
- Tags
- Multimodal-GraphRAG
- Size category
- 1K<n<10K
- Languages
- en
- Revision
- 059d3521efa7063055aa2ccea80906c264fffa33
- Last updated
- 2026-10-01
Files
12 files, 3.1 GB in total.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| annotations/benchmark.json | Data | 5.6 MB | — |
| archives/doc_pdfs.tar.zst | Data | 594.3 MB | c2b8f41b341e |
| archives/layout_content.tar.zst | Data | 7.5 MB | bbf21b76355b |
| archives/layout_images.tar.zst | Data | 995.6 MB | 6a618e6e8f6e |
| archives/layout_text_images.tar.zst | Data | 423.5 MB | 557eec222a78 |
| archives/loaded_info.tar.zst | Data | 6.8 MB | fde0811abcf5 |
| archives/page_content.tar.zst | Data | 7.9 MB | 85060bbb01f8 |
| archives/page_images.tar.zst | Data | 1.1 GB | 15e4760c043c |
| croissant.json | Data | 25.7 KB | — |
| manifest.json | Data | 5.3 MB | — |
| README.md | Documentation | 10.1 KB | — |
| .gitattributes | Repository | 2.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- No access gate
Released by RrrealLL through its official repository on Hugging Face.