Metraipolis › archive › dc_5d35ec904504
Evaluating In-Context Learning and Retrieval Strategies for Devanagari Post-OCR Correction — arXiv:2609.21595
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.21595
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T01:27:32.254Z
- SHA-256 of the archived text
e1cfa46ebda207000991c0fbc336979e45a30bdac8e09fa0a5a3c3a35f16990d- Stable URL of this page
https://metraipolis.com/documento/dc_5d35ec904504
How to cite
Evaluating In-Context Learning and Retrieval Strategies for Devanagari Post-OCR Correction — arXiv:2609.21595 — arXiv (en), https://arxiv.org/abs/2609.21595 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5d35ec904504 (Metraipolis archive dc_5d35ec904504, sha256 e1cfa46ebda207000991c0fbc336979e45a30bdac8e09fa0a5a3c3a35f16990d)
BibTeX
@misc{dc5d35ec904504,
title = {Evaluating In-Context Learning and Retrieval Strategies for Devanagari Post-OCR Correction — arXiv:2609.21595},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.21595},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5d35ec904504; sha256 e1cfa46ebda207000991c0fbc336979e45a30bdac8e09fa0a5a3c3a35f16990d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21595. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In-context learning using Large Language Models (LLMs) offers a compelling path to training-free post-OCR correction, yet its effectiveness for Devanagari script remains entirely unexplored. We present the first systematic evaluation of LLMs (3B-32B) for post-OCR correction in Hindi and Marathi, comparing three in-context example retrieval strategies: domain-random selection, dense semantic retrieval, and our proposed CharBM25, which retrieves examples by character n-gram BM25 similarity over OCR inputs to target shared error patterns with the test sentence. Across a 20,000-sentence benchmark spanning five news domains, retrieval strategy is the decisive factor in correction quality: CharBM25 outperforms domain-random selection by 2.8-4.0pp absolute WER on Hindi and 2.9-3.8pp on Marathi, using character trigrams, which consistently outperform bigrams and unigrams. Scale dominates performance: Gemma-3-27B achieves WER reductions of 55.0% for Hindi and 33.3% for Marathi under CharBM25-5. Few-shot gains are capacity-gated: models below 8B do not reliably improve over the OCR baseline, and on Marathi the smallest models (3B) degrade more sentences than they improve. Marathi is persistently harder to correct than Hindi across all scales, reflecting its greater morphological complexity. These findings establish CharBM25 as an effective, GPU-free retrieval strategy that matches or exceeds dense retrieval at negligible computational cost, and show that combining it with a general-purpose LLM of 12B+ parameters delivers reliable, training-free Devanagari post-OCR correction without task-specific fine-tuning. Dataset: this https URL Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-21 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.21595 — 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 2026-10-10T01:00:30.290Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).