AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

MEH: A Multi-Style Dataset and Toolkit for Advancing Egyptian Hieroglyph Recognition

Maksim Golyadkin, Valeria Rubanova, Aleksandr Utkov, Dmitry Nikolotov, Ilya Makarov

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Multimodal LLM benchmarks · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add synthetic word data (diffusion-rendered) to render-only pretraining (ablation)TrOCR Large · MEH real test setCER (%): 61.0 -> 50.7 (before fine-tuning); 33.8 -> 32.2 (after)Improved
add synthetic clause data on top of synthetic words (ablation)TrOCR Large · MEH real test setCER (%): 50.7 -> 46.0 (before fine-tuning); 32.2 -> 31.0 (after)Improved
add real training data after synthetic pretraining (ablation)TrOCR Large · MEH real test setCER (%): 46.0 -> 32.1 (before fine-tuning); 31.0 -> 30.6 (after)Improved
add synthetic word data (ablation)TrOCR Base · MEH real test setCER (%): 69.1 -> 55.1 (before fine-tuning); 40.2 -> 34.8 (after)Improved
add synthetic word data (ablation)MSB · MEH real test setCER (%): 117.3 -> 81.0 (before fine-tuning); 79.7 -> 62.7 (after)Improved
add synthetic clause data on top of synthetic words (ablation)MSB · MEH real test setCER (%): 81.0 -> 80.5 (before fine-tuning); 62.7 -> 62.9 (after)No clear effect
add real training data to MSB pretraining (ablation)MSB · MEH real test setCER (%): 80.5 -> 63.2 (before fine-tuning); 62.9 -> 57.4 (after)Mixed
add real data to TrOCR Small after synthetic clause pretraining (ablation)TrOCR Small · MEH real test setCER (%): 59.8 -> 52.9 (before fine-tuning); 42.2 -> 42.9 (after)Mixed
replace default tokenizer with extended tokenizer (one token per Gardiner code) (ablation)TrOCR Large · ClauserendCER (%): 5.6 -> 3.0Improved
replace default tokenizer with extended tokenizer (ablation)TrOCR Small · WordrendCER (%): 8.4 -> 9.0Mixed
scale TrOCR from Small to Large (ablation)TrOCR · Printed held-out styleCER (%): 31.5 -> 18.9Improved
scale TrOCR from Small to Large (ablation)TrOCR · Physical held-out styleCER (%): 68.1 -> 45.3Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy