Punctuation restoration Model and Spacing Model for Korean Ancient Document

TL;DR

This paper introduces the first models for punctuation and spacing restoration in Korean ancient documents, enabling better interpretation and translation of historical texts with high accuracy and low resource requirements.

Contribution

The paper presents novel models specifically designed for Korean historical texts, addressing data differences from Chinese models and achieving high performance with low VRAM usage.

Findings

Punctuation restoration F1 score of 0.84

Spacing model achieved 0.96 score

Models enable inference on low-performance GPUs

Abstract

In Korean ancient documents, there is no spacing or punctuation, and they are written in classical Chinese characters. This makes it challenging for modern individuals and translation models to accurately interpret and translate them. While China has models predicting punctuation and spacing, applying them directly to Korean texts is problematic due to data differences. Therefore, we developed the first models which predict punctuation and spacing for Korean historical texts and evaluated their performance. Our punctuation restoration model achieved an F1 score of 0.84, and Spacing model achieved a score of 0.96. It has the advantage of enabling inference on low-performance GPUs with less VRAM while maintaining quite high accuracy.