Model card
DocFormerv2-Large.
Adobe Researchopen-sourceUnknown paramsMultimodal encoder with spatial-aware cross-attention
DocFormerv2: Local Features for Document Understanding. Encoder-decoder architecture exploiting local spatial features. Large variant achieves strong mAP on DocLayNet. arXiv 2306.01733 (2023).
§ 01 · Benchmarks
Every benchmark DocFormerv2-Large has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | DocLayNet | Computer Vision · Document Understanding | mAP | 84.1% | #1 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where DocFormerv2-Large actually performs.
§ 05 · Sources & freshness
Where these numbers come from.
arxiv-paper
1
result
0 of 1 rows marked verified.