Ancient languages contain rich human historical and cultural wealth. So far there has been some good advancement in applying language technologies to ancient languages such as Sumerian, Akkadian, Latin, Ancient Greek and Ancient Chinese, especially in the construction of digital language resources and resources to facilitate automatic analysis. The workshop on Ancient Language Processing aims to focus specifically on ancient languages and scripts from the emergence of writing in Mesopotamia and Egypt c. 3000 BCE to the entire world up till 800 AD. We wish to provide a recognized forum to further advance this subfield of NLP, where researchers and practitioners can meet and discuss their latest work, and exchange ideas in addressing shared epigraphical challenges in language processing across various ancient languages, such as non-Latin and non-alphabetic scripts, Right-to-Left, transliteration conventions and fragmentary texts. In addition, we propose two shared tasks: EvaCun, focusing on using LLMs for Cuneiforms, and EvaHan, addressing Named Entity Recoganition for Ancient Chinese. These tasks aim to provide an opportunity to tackle the unique challenges of ancient language processing.
Languages of interest include, but are not limited to:
Mesopotamia: Sumerian, Akkadian
Iran: Elamite, old and middle Persian
Levant: Eblaite, Amorite, Aramaic (incl. Mandaic and Syriac), Ancient Hebrew, Phynician, Ugaritic
Anatolia: Hittite, Luwian and minor Anatolian languages
Egypt: Ancient Egyptian, Coptic
Mediterranean: Linear A and B, Ancient Greek, Latin
Arabia: Ancient North Arabian, old Arabic
India: Sanskrit, Eastern Panjabi, Pali
China: Literary Chinese, Tibetan
Mesoamerica: Mayan
Japan: Old Japanese
Papers and contributions are encouraged for any work related to Natural Language Processing of Ancient Languages. Topics of interest include, but are not limited to:
Charset (Unicode)
Input method (transliteration and transcription)
Tokenization (word segmentation)
Morphological analysis (both inflectional and derivational)
Philological issues in NLP
Linguistic Linked Data supporting NLP
Syntactic analysis
Semantic analysis
Machine translation
Pre-trained models
Deep learning based NLP
Multi-lingual comparison for NLP purposes
Data mining
Knowledge extraction
Language varieties and dialects
NLP issues in the analysis of broken texts and uncertain readings
Minimal computing in NLP
We welcome three types of submissions:
-Long papers (full papers) that describe original and unpublished work in any topic area of the workshop. A long paper is limited to 8 pages for content, with unlimited number of pages for references.
-Short papers (posters) that describe either work in progress or a research proposal. They may also be in the style of a position paper that surveys and criticizes existing literature. Short papers must include clear directions for future research. Submissions of this type are limited to 4 pages for content, with unlimited number of pages for references.
-Tech report papers (for Shared Tasks) that describe work in either the EvaHan or the EvaCun shared task. They may also be in the style of a position paper that surveys and criticizes existing literature. Tech papers must include clear descriptions for their method and system performance. Submissions of this type are limited to 4 pages for content, with unlimited number of pages for references.