Metraipolis › archive › dc_a220cf32f893
WildHandBench: A Benchmark for Handwritten Text Understanding that Challenges MLLMs and Humans — arXiv:2608.22959
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.22959
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T07:45:07.725Z
- SHA-256 of the archived text
fe7dd9f7f2c1cf98c796d425bed9adc6354c3fb6a6ef6c3d706362b39c59f03d- Stable URL of this page
https://metraipolis.com/documento/dc_a220cf32f893
How to cite
WildHandBench: A Benchmark for Handwritten Text Understanding that Challenges MLLMs and Humans — arXiv:2608.22959 — arXiv (en), https://arxiv.org/abs/2608.22959 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a220cf32f893 (Metraipolis archive dc_a220cf32f893, sha256 fe7dd9f7f2c1cf98c796d425bed9adc6354c3fb6a6ef6c3d706362b39c59f03d)
BibTeX
@misc{dca220cf32f893,
title = {WildHandBench: A Benchmark for Handwritten Text Understanding that Challenges MLLMs and Humans — arXiv:2608.22959},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.22959},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a220cf32f893; sha256 fe7dd9f7f2c1cf98c796d425bed9adc6354c3fb6a6ef6c3d706362b39c59f03d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.22959. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
While the top model on OmniDocBench now reaches 96.34% overall on printed-document parsing, the ability of current models to handle challenging handwritten documents remains largely uncharacterized. Existing benchmarks focus on isolated text or formulas, overlook handwritten tables and real-world degradation, and report aggregate accuracy without explaining why models fail. We present WildHandBench, a benchmark containing 500 handwritten documents across three structures (free text, tables, formulas), four languages, and nine real-world scenarios. We introduce a Prior-Driven Error (PDE) metric that quantifies whether errors originate from language priors rather than visual evidence. Evaluating 18 state-of-the-art models together with calibrated human baselines, we find: (1) the best model achieves only 71.85% overall; (2) humans outperform all models yet the gap is narrow (77.09% vs. 71.85%); and (3) model errors are qualitatively different from human errors -- 63-91% of model errors are prior-driven versus only 49% for humans, exposing systematic reliance on language priors that conventional accuracy metrics cannot capture. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.22959 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T07:37:52.909Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).