post-ocr-cleanup

Name: post-ocr-cleanup
Author: scdenney

34stars0forks

How to install post-ocr-cleanup?

scdenney/open-science-skills/post-ocr-cleanup

$npx -y skills add scdenney/open-science-skills --skill post-ocr-cleanup

Installs into the current project.

Summary

Clean post-OCR research text. Use for correction, quality assurance, multilingual handling, uncertainty preservation, and provenance-aware cleanup.

Files · 1

View on GitHub

SKILL.md

1	# Post-OCR Text Cleanup for Research Corpora
2
3	## Instructions
4
5	### 1. Cleanup Strategy Selection
6
7	- Characterize the error-generating DGP before selecting a method. Document source language(s), era, typeface family (Fraktur, Antiqua, typewritten, handwritten), scan DPI, OCR engine, and domain jargon. Each parameter constrains which corrections are plausible and which risk introducing semantic drift.
8	- Choose between LLM correction, rule-based fixes, or a hybrid pipeline based on error type. LLM correction excels at context-dependent errors (wrong but plausible characters, broken words, missing diacritics). Rule-based fixes handle deterministic patterns (control characters, Unicode normalization, repetition artifacts, whitespace) with zero risk of content alteration. Use rule-based fixes unconditionally for these categories.
9	- Default to the hybrid approach for research corpora. Run LLM correction first on all pages, then apply deterministic rule fixes on top. This order matters: LLM correction may introduce formatting artifacts that rule fixes clean up, while the reverse order wastes rule-fix effort on text the LLM will rewrite (Machidon & Machidon 2025).
10	- Pilot-test LLM correction per language before corpus-wide deployment. LLM post-correction effectiveness is highly language-dependent: English achieves 7-58% CER reduction across open models, while Finnish shows negative or near-zero improvement across the same model set (Kanerva et al. 2025). Never assume cross-language transferability.
11	- Consider whether correction is needed at all. Define the quality threshold before choosing a strategy. The Hill & Hengchen 70-80% quality band (reported in van Strien et al. 2020) marks the critical threshold below which most downstream NLP tasks perform poorly; above 80% quality many tasks (e.g., topic modeling) tolerate residual noise. If the downstream analysis sits comfortably above this band, the risk of correction-introduced errors may outweigh the benefit.
12
13	### 2. LLM-Based Correction
14
15	- For most pages, use a small text-only model. The correction input is already text; image understanding is not needed for well-OCR'd pages. A 7-13B parameter model with 4-bit quantization fits in ~4-20GB VRAM and runs on a single GPU. Larger fp16 models (e.g., Llama-3.1-70B at fp16 yielding ~42% CER reduction vs ~39% at 4-bit) gain 2.5-4.7pp but require roughly 3x the memory (132GB vs 43GB) and often a second GPU (Kanerva et al. 2025).
16	- For severely degraded pages, use multimodal correction. Feeding both the original page image and the OCR text to a correction model can achieve below 1% CER on degraded documents, but doubles GPU cost (Greif et al. 2025). Reserve this for flagged pages, not routine processing.
17	- Write tight correction prompts. Instruct the model to "fix clear OCR mistakes only: wrong characters, broken words, garbled punctuation, repetition artifacts. Do not translate, modernize, or add anything. Output the corrected text only." Loose prompts invite hallucination.
18	- Supply socio-cultural context in the prompt. Including document era, publication type, language register, and genre (e.g., "The text is from an English newspaper in the 1800s") meaningfully reduces CER beyond generic correction prompts — the top-performing CLOCR-C configuration achieved over 60% CER reduction on the NCSE dataset using a modular prompt that combines expert framing, recovery instructions, publication context, text-type context, and anti-overgeneration instructions (Bourne 2024). Misleading or mismatched context degrades performance, so use the real document metadata.
19	- Add language-specific instructions. For Polish, explicitly mention diacritics restoration (ą, ć, ę, ł, ń, ó, ś, ź, ż). For Korean, mention hangul integrity and hanja preservation. The correction model needs to know which character set to favor.
20	- Mitigate hallucination with constrained decoding. Constrained decoding techniques — beam search with CER-based re-ranking, sequence-level similarity re-ranking, and token-level Constrained Beam Search that interpolates the model's distribution with a character-similarity distribution — enforce fidelity between input and output and prevent plausible-but-fabricated substitutions (Sastre et al. 2025). Prefer token-level CBS with dynamic α if model logits are accessible; otherwise fall back to beam search with CER re-selection. This matters because WER can worsen even when CER improves: fine-tuning alone in Sastre et al. left CER roughly flat (0.314→0.321) while WER jumped from 0.633 to 0.821, a failure mode constrained decoding directly addresses.
21	- Use worked prompt templates and a provenance schema. See `reference/prompt-templates-and-schema.md` for a minimal constrained-decoding-friendly base

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

post-ocr-cleanup

34stars0forks

View on GitHub

How to install post-ocr-cleanup?

scdenney/open-science-skills/post-ocr-cleanup

$npx -y skills add scdenney/open-science-skills --skill post-ocr-cleanup

Installs into the current project.

Summary

Clean post-OCR research text. Use for correction, quality assurance, multilingual handling, uncertainty preservation, and provenance-aware cleanup.

Files · 1

View on GitHub

SKILL.md

1	# Post-OCR Text Cleanup for Research Corpora
2
3	## Instructions
4
5	### 1. Cleanup Strategy Selection
6
7	- Characterize the error-generating DGP before selecting a method. Document source language(s), era, typeface family (Fraktur, Antiqua, typewritten, handwritten), scan DPI, OCR engine, and domain jargon. Each parameter constrains which corrections are plausible and which risk introducing semantic drift.
8	- Choose between LLM correction, rule-based fixes, or a hybrid pipeline based on error type. LLM correction excels at context-dependent errors (wrong but plausible characters, broken words, missing diacritics). Rule-based fixes handle deterministic patterns (control characters, Unicode normalization, repetition artifacts, whitespace) with zero risk of content alteration. Use rule-based fixes unconditionally for these categories.
9	- Default to the hybrid approach for research corpora. Run LLM correction first on all pages, then apply deterministic rule fixes on top. This order matters: LLM correction may introduce formatting artifacts that rule fixes clean up, while the reverse order wastes rule-fix effort on text the LLM will rewrite (Machidon & Machidon 2025).
10	- Pilot-test LLM correction per language before corpus-wide deployment. LLM post-correction effectiveness is highly language-dependent: English achieves 7-58% CER reduction across open models, while Finnish shows negative or near-zero improvement across the same model set (Kanerva et al. 2025). Never assume cross-language transferability.
11	- Consider whether correction is needed at all. Define the quality threshold before choosing a strategy. The Hill & Hengchen 70-80% quality band (reported in van Strien et al. 2020) marks the critical threshold below which most downstream NLP tasks perform poorly; above 80% quality many tasks (e.g., topic modeling) tolerate residual noise. If the downstream analysis sits comfortably above this band, the risk of correction-introduced errors may outweigh the benefit.
12
13	### 2. LLM-Based Correction
14
15	- For most pages, use a small text-only model. The correction input is already text; image understanding is not needed for well-OCR'd pages. A 7-13B parameter model with 4-bit quantization fits in ~4-20GB VRAM and runs on a single GPU. Larger fp16 models (e.g., Llama-3.1-70B at fp16 yielding ~42% CER reduction vs ~39% at 4-bit) gain 2.5-4.7pp but require roughly 3x the memory (132GB vs 43GB) and often a second GPU (Kanerva et al. 2025).
16	- For severely degraded pages, use multimodal correction. Feeding both the original page image and the OCR text to a correction model can achieve below 1% CER on degraded documents, but doubles GPU cost (Greif et al. 2025). Reserve this for flagged pages, not routine processing.
17	- Write tight correction prompts. Instruct the model to "fix clear OCR mistakes only: wrong characters, broken words, garbled punctuation, repetition artifacts. Do not translate, modernize, or add anything. Output the corrected text only." Loose prompts invite hallucination.
18	- Supply socio-cultural context in the prompt. Including document era, publication type, language register, and genre (e.g., "The text is from an English newspaper in the 1800s") meaningfully reduces CER beyond generic correction prompts — the top-performing CLOCR-C configuration achieved over 60% CER reduction on the NCSE dataset using a modular prompt that combines expert framing, recovery instructions, publication context, text-type context, and anti-overgeneration instructions (Bourne 2024). Misleading or mismatched context degrades performance, so use the real document metadata.
19	- Add language-specific instructions. For Polish, explicitly mention diacritics restoration (ą, ć, ę, ł, ń, ó, ś, ź, ż). For Korean, mention hangul integrity and hanja preservation. The correction model needs to know which character set to favor.
20	- Mitigate hallucination with constrained decoding. Constrained decoding techniques — beam search with CER-based re-ranking, sequence-level similarity re-ranking, and token-level Constrained Beam Search that interpolates the model's distribution with a character-similarity distribution — enforce fidelity between input and output and prevent plausible-but-fabricated substitutions (Sastre et al. 2025). Prefer token-level CBS with dynamic α if model logits are accessible; otherwise fall back to beam search with CER re-selection. This matters because WER can worsen even when CER improves: fine-tuning alone in Sastre et al. left CER roughly flat (0.314→0.321) while WER jumped from 0.633 to 0.821, a failure mode constrained decoding directly addresses.
21	- Use worked prompt templates and a provenance schema. See `reference/prompt-templates-and-schema.md` for a minimal constrained-decoding-friendly base

People who liked this also viewed

6 picks

Type