A Context-Contrastive Inference Approach To Partial Diacritization

TL;DR

This paper presents a novel context-contrastive approach to partial diacritization in Arabic, demonstrating that selective diacritization can enhance readability and proposing new metrics and a Transformer-based model for this task.

Contribution

It introduces CCPD, a new method for partial diacritization that leverages context comparison, along with novel quality indicators and a Transformer-based model called TD2.

Findings

Partially diacritized text can be easier to read than fully diacritized or plain text.

CCPD effectively identifies characters to diacritize by comparing context-aware and context-free inferences.

TD2 shows distinct performance characteristics on new quality indicators compared to existing systems.

Abstract

Diacritization plays a pivotal role in improving readability and disambiguating the meaning of Arabic texts. Efforts have so far focused on marking every eligible character (Full Diacritization). Comparatively overlooked, Partial Diacritzation (PD) is the selection of a subset of characters to be marked to aid comprehension where needed. Research has indicated that excessive diacritic marks can hinder skilled readers -- reducing reading speed and accuracy. We conduct a behavioral experiment and show that partially marked text is often easier to read than fully marked text, and sometimes easier than plain text. In this light, we introduce Context-Contrastive Partial Diacritization (CCPD) -- a novel approach to PD which integrates seamlessly with existing Arabic diacritization systems. CCPD processes each word twice, once with context and once without, and diacritizes only the characters with disparities between the two inferences. Further, we introduce novel indicators for measuring partial diacritization quality, essential for establishing this as a machine learning task. Lastly, we introduce TD2, a Transformer-variant of an established model which offers a markedly different performance profile on our proposed indicators compared to all other known systems.