aiqbench analysis
TechPulse take: This research introduces MUDIDI, a two-stage framework that leverages Large Language Models to effectively digitize complex, multilingual dictionaries into machine-readable formats. By benchmarking against OCR and Vision Language Models, the study demonstrates LLM superiority in handling diverse writing systems while releasing a new dataset of 30 public-domain dictionaries to support further evaluation. The work provides actionable guidelines for improving digitization quality, particularly when supplementing models with contextual dictionary information.
MUDIDI: A Two-Stage Framework for Multilingual Dictionary Digitization with Language Models
arXiv:2606.09435v1 Announce Type: new Abstract: Multilingual dictionaries are among the most valuable documentary resources for low-resource and endangered languages, yet many remain available only as scans. For many decades, their digitization and conversion into a machine-readable format was nearly impossible due to language-specific scripts, complex multi-column layouts full of entries with abbreviations and cross-references. Recent vision-language models offer a promising solution, but it is unclear how well they preserve characters, markup, and process lexicographic structure. We introduce MUDIDI, a two-stage framework for multi-lingual dictionary digitization. Stage One evaluates the quality of character recognition and markup preservation; Stage Two focuses on dictionary entry segmentation with subsequent mapping into a machine-readable lexicographic schema, SIL's Multi-Dictionary Formatter. We also release a dataset that consists of human-annotated lexicographic entries collected from 30 public-domain dictionaries featuring diverse writing systems, language families, and formats. We benchmark OCR systems, general-purpose Large Language Models (LLMs), and Vision Language Models (VLMs) on the dataset, demonstrating superior performance of LLMs across most writing systems and languages in both stages, and provide practical guidelines on improving the results for more challenging scenarios. Finally, we show that supplementing additional information, such as dictionary introduction, to the LLMs can improve the quality of the digitized dictionary. Github: https://github.com/DavidSamuell/MUDIDI-Pipeline-for-Digitization-of-Multilingual-Dictionary/
Abstract:Multilingual dictionaries are among the most valuable documentary resources for low-resource and endangered languages, yet many remain available only as scans. For many decades, their digitization and conversion into a machine-readable format was nearly impossible due to language-specific scripts, complex multi-column layouts full of entries with abbreviations and cross-references. Recent vision-language models offer a promising solution, but it is unclear how well they preserve characters, markup, and process lexicographic structure. We introduce MUDIDI, a two-stage framework for multi-lingual dictionary digitization. Stage One evaluates the quality of character recognition and markup preservation; Stage Two focuses on dictionary entry segmentation with subsequent mapping into a machine-readable lexicographic schema, SIL's Multi-Dictionary Formatter. We also release a dataset that consists of human-annotated lexicographic entries collected from 30 public-domain dictionaries featuring diverse writing systems, language families, and formats. We benchmark OCR systems, general-purpose Large Language Models (LLMs), and Vision Language Models (VLMs) on the dataset, demonstrating superior performance of LLMs across most writing systems and languages in both stages, and provide practical guidelines on improving the results for more challenging scenarios. Finally, we show that supplementing additional information, such as dictionary introduction, to the LLMs can improve the quality of the digitized dictionary. Github: this https URL
Computation and Language (cs.CL)
arXiv:2606.09435 [cs.CL]
(or arXiv:2606.09435v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2606.09435 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 297 words
Related coverage
MUDIDI: A Two-Stage Framework for Multilingual Dictionary Digitization with Language Models
arXiv · Sep 3, 2026
Evaluate Clinical ASR Models Faster with Agent Skills and NVIDIA Nemotron Speech
NVIDIA · Jun 9, 2026
Digital States 2026: AI Moves From Experiment to Infrastructure
govtech.com · Sep 24, 2026
Artificial Intelligence-Assisted Digital Inventory of Cultural Heritage & Traditional Knowledge: Case for Indonesian Open Digital Library of Culture
arXiv · Sep 9, 2026