Here is a list of upcoming and past events related to the RWALT project:
Upcoming Events
No upcoming events
Past Events
RWALT @ RSA2025
Introducing the “Representing Women Authorship in the Latin Treebanks” Project
3.22.2025 @ 4:30PM Eastern. Boston, MA
RSA2025 Roundtable on “Teaching Neo-Latin Texts by Women: Strategies and Innovations in Latin Education”
Link
Abstract: In this roundtable contribution, I introduce proposed work on developing an annotated dataset—specifically a Universal Dependencies-compatible treebank—of Latin texts written by women. Despite the availability of five UD Latin treebanks, i.e. textual data structures consisting of word-for-word annotations of sentences, there are at present no works represented in these collections written by women. This data curation and digitization project takes initial steps in improving gender representation in available Latin datasets and so by extension in available applications, tools, and resources trained on these datasets. This roundtable contribution 1. describes the project; 2. explains the benefits of this kind of data-specific digitization to other computational projects designed for supporting Latin teaching; 3. shows how the project fits in with related digitization projects (e.g. Project Nota); and 4. explains opportunities for Latin teachers and students to contribute to the project regardless of technical background.
RWALT @ RELICS
Initial Steps toward a Linguistic Dataset of Latin Texts Written by Women
3.11.2025 @ 10AM Eastern. Online.
RELICS Roundtable on “Women as Authors of Latin Literature”
Link
Abstract: Treebanks are a fundamental linguistic data structure used for training text analysis pipelines that enable computational lemmatization, part-of-speech tagging, and named entity recognition, among many other tasks. While training a collection of such pipelines for the Latin language (LatinCy), I became aware that the available Latin treebanks, while varied with respect to qualities like chronology, geography, and genre, failed to include any texts written by women. In an effort to redress this dataset imbalance, I launched the “Representing Women’s Authorship in the Latin Treebanks” (RWALT) project in the Fall 2024 semester. RWALT currently has three students curating metadata about women Latin authors as well as collecting (and correcting) semantic and syntactic annotations based on their writings that will be used as the basis for a new thematic Latin treebank. This short talk details the early stages of our team’s work on the RWALT project, with specific attention to our annotation workflow. Through participation in this roundtable, we are particularly interested in hearing from experts on the topic about what principles of selection and data design decisions they would recommend for the project so that our development work will be maximally beneficial to the larger research community.