진학프로석·박사 채용 접수 플랫폼

This position is offered as part of the BREAKTHROUGH project, that received support from the French government and is managed by the National Research Agency (ANR), under the France 2030 program as part CaeSAR project, reference "ANR-23-EXES-0001” and the Normandy Region. Led by Professor Pierre Larrivée, the BREAKTHOUGH project will be running from 1 April 2026 to 30 September 2026. The aim of the project is to optimise semi-automatic annotation of historical texts in Romance languages (medieval and early modern Norman French and Occitan).

Position Overview

The researcher will contribute to the development of a semi-automated workflow for high-quality annotation of corpora in historical Romance varieties and dissemination of best practices among historical linguistics in France and internationally. He/She will engage in production of annotated corpora and training of models for low-resourced historical Romance varieties.

Key Responsibilities

1. 1. Constitution of annotated corpora
   
   • • Produce syntactically annotated calibrated corpora of historical texts in medieval and early modern Norman French and Occitan, annotated in Universal Dependencies framework (treebanks).
   • • Explore the corpora to quantify the evolution of syntactic phenomena (e.g. loss of pro-drop) across the period.
   • • Train and evaluate annotation models based on these corpora.
   • • Make the corpora available for consultation via the CRISCO lab TXM portal and the Universal Dependencies collection.
     
   • 
     
2. 2. Development and dissemination of best practices for high-quality annotation
   
   • • Produce documentation on the use of tools and methods for annotation of historical Romance corpora.
   • • Deliver at least six training sessions for colleagues and students on the annotation of historical corpora, in France and internationally.
     
   • 
     
3. 3. Collaboration with project partners and organisation of research events
   
   • • Maintain close communication with the project partners to exchange research results and best practices.
   • • Organise online meetings and visits for the project partners.
   • • Organise at least one international conference or study day on historical corpus linguistics at the University of Caen.
     
   • 
     
4. Training of students
   
   • • Supervise interns in treebank annotation and natural language processing.