Methodology & Architecture

1. System Architecture: Three-Tier Collaborative Routing

Preserving digitized historical archives requires mitigating transcription degradation without incurring unsustainable computational inference costs or inducing catastrophic hallucination. Naive application of Large Language Models (LLMs) across entire archival collections results in prohibitive expenses, vocabulary modernization, and frequent regression of correctly transcribed rare historical entities.

Our architecture introduces a three-tier routing paradigm that dynamically allocates each text segment to one of three optimal treatment paths based on learned degradation profiles and empirical confidence safeguards:

Figure 1: Collaborative Human-LLM Post-OCR Correction Architecture with Dynamic Confidence Routing and Post-Correction Safeguard Loop.

Class 0: Bypass Route (Direct to Archive)

Segments identified as already clean or high-confidence bypass LLM correction completely. Saves 100% token cost and eliminates any risk of stylistic hallucination or diacritic corruption.

Class 1: Automated LLM Correction

Moderately degraded segments are dispatched to historical prompting pipelines (Gemini / GPT / Llama) with few-shot archival examples and strict orthographic grounding constraints.

Class 2: Priority Human Verification (<5%)

Severe segmentation failures, empty OCR blocks, or segments intercepted by the post-correction safeguard are routed to professional archivists under a strict <5% human time budget.

2. Mathematical Formulation & Quality Metrics

Let D = {d₁, …, d_n} denote a digitized archival corpus. Each document segment d_i possesses an original raw OCR transcription d_i^raw, a candidate machine correction d_i^corr, and a ground-truth transcription d_i^human.

We quantify correction quality via the Character Error Rate (CER) reduction benefit Δ_i:

Δ_i = CER(d_i^raw, d_i^human) − CER(d_i^corr, d_i^human)

Positive Benefit (Δ_i > 0): The machine correction successfully lowered transcription noise and improved readability.
Negative Regression (Δ_i < 0): The LLM degraded accuracy, hallucinated false cognates, or altered archaic syntax. The post-correction safeguard is designed to catch these regressions.

3. Feature Extraction & LASSO Degradation Modeling

Prior to invoking expensive generative models, the router extracts 54 lightweight features capturing multi-faceted document degradation without requiring ground truth annotations:

Feature Family	Extracted Indicators	Routing Rationale
Confidence Metrics	Mean token confidence, lowest quartile confidence, token variance	Low confidence triggers LLM or human verification
Lexical & Orthographic	Out-of-vocabulary ratio against 18th/19th c. Swiss French lexicons, symbol density	Distinguishes historical spelling from OCR garbage characters
Perplexity & Language Model	CamemBERT cross-entropy loss, token log-likelihoods	Captures ungrammatical syntax disruptions caused by broken line breaks
Typographic & Layout	Bounding box aspect ratio, average word length, uppercase-to-lowercase transitions	Identifies broken column cuts and hyphenation splits

Figure 2: LASSO Feature Selection Coefficients identifying the strongest predictive drivers of post-OCR correction benefit (Δ_i).

4. Safeguard Routing Curves & Verification Budget

A critical vulnerability in all-LLM correction pipelines is the silent corruption of historical names and rare dates. To eliminate this risk, our framework deploys a lightweight Post-Correction Safeguard classifier evaluated on the pair (d_i^raw, d_i^corr):

Computes normalized Levenshtein distance, word count divergence, and proper noun preservation.
If the safeguard predicts that the candidate correction introduced regressions (CER(d_i^corr) > CER(d_i^raw)), the text is intercepted prior to writing to the database and escalated directly to human review.

Figure 3: Empirical Safeguard Routing Boundaries balancing Character Error Rate reduction against the strict human verification budget (<5%).

Key Architectural Takeaways

Cost Reduction: Bypassing 60%+ of clean segments cuts API inference expenditure by more than half.
Guaranteed Quality: The dual-gate architecture (Router + Safeguard) prevents hallucinated modernizations from ever corrupting the archival repository.
Human Efficiency: Prioritizing human review exclusively on high-uncertainty and safeguard-flagged segments maximizes archival accuracy per reviewer hour.