Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Quid est VERITAS? A Modular Framework for Archival Document Analysis

Leonardo Bassanini, Ludovico Biancardi, Alfio Ferrara, Andrea Gamberini, Sergio Picascia, Folco Vaglienti

Abstract

The digitisation of historical documents has traditionally been conceived as a process limited to character-level transcription, producing flat text that lacks the structural and semantic information necessary for substantive computational analysis. We present VERITAS (Vision-Enhanced Reading, Interpretation, and Transcription of Archival Sources), a modular, model-agnostic framework that reconceptualises digitisation as an integrated workflow encompassing transcription, layout analysis, and semantic enrichment. The pipeline is organised into four stages - Preprocessing, Extraction, Refinement, and Enrichment - and employs a schema-driven architecture that allows researchers to declaratively specify their extraction objectives. We evaluate VERITAS on the critical edition of Bernardino Corio's Storia di Milano, a Renaissance chronicle of over 1,600 pages. Results demonstrate that the pipeline achieves a 67.6% relative reduction in word error rate compared to a commercial OCR baseline, with a threefold reduction in end-to-end processing time when accounting for manual correction. We further illustrate the downstream utility of the pipeline's output by querying the transcribed corpus through a retrieval-augmented generation system, demonstrating its capacity to support historical inquiry.

Quid est VERITAS? A Modular Framework for Archival Document Analysis

Abstract

The digitisation of historical documents has traditionally been conceived as a process limited to character-level transcription, producing flat text that lacks the structural and semantic information necessary for substantive computational analysis. We present VERITAS (Vision-Enhanced Reading, Interpretation, and Transcription of Archival Sources), a modular, model-agnostic framework that reconceptualises digitisation as an integrated workflow encompassing transcription, layout analysis, and semantic enrichment. The pipeline is organised into four stages - Preprocessing, Extraction, Refinement, and Enrichment - and employs a schema-driven architecture that allows researchers to declaratively specify their extraction objectives. We evaluate VERITAS on the critical edition of Bernardino Corio's Storia di Milano, a Renaissance chronicle of over 1,600 pages. Results demonstrate that the pipeline achieves a 67.6% relative reduction in word error rate compared to a commercial OCR baseline, with a threefold reduction in end-to-end processing time when accounting for manual correction. We further illustrate the downstream utility of the pipeline's output by querying the transcribed corpus through a retrieval-augmented generation system, demonstrating its capacity to support historical inquiry.

Paper Structure

This paper contains 18 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Transcription of Historical Documents
  4. Integrated Digitisation Pipelines
  5. Methodology
  6. Preprocessing
  7. Extraction
  8. Refinement
  9. Enrichment
  10. Evaluation
  11. Case Study
  12. Evaluation Methodology
  13. Results
  14. Downstream Application
  15. Setup
  16. ...and 3 more sections

Figures (4)

  • Figure 1: VERITAS architecture diagram illustrating the four stages and their inputs and outputs.
  • Figure 2: Preprocessing operations applied to a page from the critical edition of Corio's Storia di Milano. Left: the original colour scan, captured in landscape orientation with portions of the adjacent page visible. Right: the result after rotation correction, page detection, grayscale conversion, and adaptive thresholding.
  • Figure 3: The difference in output between the Specialised Vision-Language Model and the General-Purpose Vision-Language Model processing paths.
  • Figure 4: Examples of Enrichment operations.