analiza aiqbench
This research introduces ARI, a framework that enhances historical document restoration by integrating retrieval-augmented generation (RAG) with large language models to recover external knowledge for proper nouns. The approach demonstrates significant performance gains over standard masked language modeling baselines when applied to deteriorated Korean historical texts. By effectively combining implicit model knowledge with retrieved context, this method offers a practical tool for accelerating the analysis of damaged archival records without relying on quantum or general-purpose AI features as its core novelty.
Leveraging External Knowledge for Historical Document Restoration via Retrieval-Augmented Large Language Models
arXiv:2607.21936v1 Announce Type: new Abstract: Historical documents act as invaluable knowledge archives but often suffer from illegibility due to physical deterioration and damage. While existing restoration methods based on masked language modeling effectively utilize local context, they struggle to restore named entities that require external historical knowledge. To address this limitation, we introduce a novel framework for historical document restoration that leverages large language models with retrieval-augmented generation (RAG). By combining the implicit knowledge of pre-trained LLMs with explicitly retrieved external context, our model ARI effectively mitigates the challenge of inferring context-dependent proper nouns. Extensive experiments on Korean historical documents demonstrate that our approach significantly outperforms baselines, achieving substantial gains in restoring both general characters and named entities. Furthermore, comprehensive evaluations including expert assessments confirm that ARI serves as a practical tool for domain experts, promising to accelerate the analysis of historical records.
Abstract:Historical documents act as invaluable knowledge archives but often suffer from illegibility due to physical deterioration and damage. While existing restoration methods based on masked language modeling effectively utilize local context, they struggle to restore named entities that require external historical knowledge. To address this limitation, we introduce a novel framework for historical document restoration that leverages large language models with retrieval-augmented generation (RAG). By combining the implicit knowledge of pre-trained LLMs with explicitly retrieved external context, our model ARI effectively mitigates the challenge of inferring context-dependent proper nouns. Extensive experiments on Korean historical documents demonstrate that our approach significantly outperforms baselines, achieving substantial gains in restoring both general characters and named entities. Furthermore, comprehensive evaluations including expert assessments confirm that ARI serves as a practical tool for domain experts, promising to accelerate the analysis of historical records.
Computation and Language (cs.CL)
arXiv:2607.21936 [cs.CL]
(or arXiv:2607.21936v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2607.21936 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
https://doi.org/10.18653/v1/2026.findings-acl.2148 https://doi.org/10.18653/v1/2026.findings-acl.2148 Focus to learn more DOI(s) linking to related resources
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.
Text integral sub licența deschisă a publicației · 244 cuvinte
Subiecte conexe
Inference Cost Attacks for Retrieval-Augmented Large Language Models
arXiv · 3 iun. 2026
A Survey on Knowledge-Oriented Retrieval-Augmented Generation
arXiv · 21 iul. 2026
SKILL-RAG: Self-Knowledge Induced Learning and Filtering for Retrieval-Augmented Generation
arXiv · 24 aug. 2026
Digital States 2026: AI Moves From Experiment to Infrastructure
govtech.com · 24 sept. 2026