Model card
TABLET.
Anonymous (arXiv 2025)open-sourceUnknown paramsDual Transformer encoders; encoder-only architecture; row/column splitting as sequence labeling
Split-Merge top-down model. Jun 2025. Achieves TEDS 96.79, TEDS-Struct 97.67 on PubTabNet validation set.
§ 01 · Benchmarks
Every benchmark TABLET has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | pubtabnet | Computer Vision · Table Recognition | teds-all-samples | 96.8% | #3 | 2025-06-08 | source ↗ |
| 02 | pubtabnet | Computer Vision · Table Recognition | teds-struct | 97.7% | #4 | 2025-06-08 | 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.
§ 03 · Papers
1 paper with results for TABLET.
- 2025-06-08· Computer Vision· 2 results
TABLET: Table Structure Recognition using Encoder-only Transformers
§ 05 · Sources & freshness
Where these numbers come from.
arxiv
2
results
2 of 2 rows marked verified.