LogogramNLP: Comparing Visual and Textual Representations of Ancient Logographic Writing Systems for NLP

TL;DR

This paper introduces LogogramNLP, a benchmark for analyzing ancient logographic languages using visual and textual data, showing visual methods can outperform text-based ones for certain NLP tasks.

Contribution

It presents the first benchmark for NLP on ancient logographic languages, comparing visual and textual representations and demonstrating the potential of visual processing.

Findings

Visual representations outperform textual ones on some tasks

The benchmark includes datasets for classification, translation, and parsing

Visual processing can unlock cultural heritage data for NLP

Abstract

Standard natural language processing (NLP) pipelines operate on symbolic representations of language, which typically consist of sequences of discrete tokens. However, creating an analogous representation for ancient logographic writing systems is an extremely labor intensive process that requires expert knowledge. At present, a large portion of logographic data persists in a purely visual form due to the absence of transcription -- this issue poses a bottleneck for researchers seeking to apply NLP toolkits to study ancient logographic languages: most of the relevant data are images of writing. This paper investigates whether direct processing of visual representations of language offers a potential solution. We introduce LogogramNLP, the first benchmark enabling NLP analysis of ancient logographic languages, featuring both transcribed and visual datasets for four writing systems along with annotations for tasks like classification, translation, and parsing. Our experiments compare systems that employ recent visual and text encoding strategies as backbones. The results demonstrate that visual representations outperform textual representations for some investigated tasks, suggesting that visual processing pipelines may unlock a large amount of cultural heritage data of logographic languages for NLP-based analyses.