LogogramNLP: Comparing Visual and Textual Representations of Ancient Logographic Writing Systems for NLP
This work addresses a problem for researchers studying ancient logographic languages by enabling NLP analysis of visual data, though it is incremental as it builds on existing visual and text encoding methods.
The paper tackled the bottleneck of applying NLP to ancient logographic languages, which often lack transcriptions, by introducing LogogramNLP, a benchmark with visual and textual datasets for four writing systems, and found that visual representations outperformed textual ones in some tasks.
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.