MMLongBench-Doc
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multimodal Long-document Understanding | MMLongBench-Doc 1.0 (test) | Reports Score46.7 | 12 | |
| Multimodal VQA | MMLongBench-Doc 2K Packed | Accuracy (Residential)55.66 | 11 | |
| Question Answering | MMLongBench-Doc Multi-page (MP) | EM24.18 | 10 | |
| Document Retrieval | MMLongBench-Doc Multi-page (MP) | Recall61.38 | 10 | |
| Question Answering | MMLongBench-Doc Single-page | EM29.34 | 10 | |
| Document Retrieval | MMLongBench-Doc Single-page (SP) | Top-1 Accuracy75.77 | 10 | |
| Long-context document understanding | MMLongBench-Doc 32K context slice | MMLongBench-Doc (32K) Score45 | 4 | |
| Long-context document understanding | MMLongBench-Doc 16K context slice | Score48.57 | 4 |
Showing 8 of 8 rows