How Far Can Synthetic Data Take Thai OCR?
Synthetic-only training adapts PaddleOCR-VL into Wayu-Paxa-OCR-Zero, cutting Thai printed-page CER from 6.64% to 1.24% without real labels.
The study disentangles which factors of synthetic OCR data transfer to real Thai documents, finding typeface diversity, 2D structure, and real handwriting glyphs matter most. Using 45,723 synthetic pages, the authors adapt the 0.9B-parameter PaddleOCR-VL-1.6 into Wayu-Paxa-OCR-Zero, reducing median CER from 6.64% to 1.24% on printed pages and from 74.87% to 20.55% on handwriting. The model outperforms Typhoon OCR v1 7B on all five evaluation sets.
- Wayu-Paxa-OCR-Zero adapted from PaddleOCR-VL-1.6 using 45,723 synthetic pages
- Printed-page median CER drops from 6.64% to 1.24%; handwriting from 74.87% to 20.55%
- Outperforms Typhoon OCR v1 7B on all five evaluation sets
- Typeface diversity, 2D structure, and real handwriting glyphs drive transfer
From the original article
1 min read at huggingface.co
We investigate what makes synthetic OCR supervision transfer to real Thai documents and use the resulting insights to build Wayu-Paxa-OCR-Zero, a Thai OCR model adapted without OCR labels from real Thai document pages. Synthetic data provide exact labels at scale, but "realism" conflates source domain, page context, typography, spatial structure, and glyph variation.
By Kunat Pipatanakul · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.