Automatic generation of section headings in legal texts

Niveau : Master

Période : 6 months

Location: Nantes University, Laboratory of Digital Sciences of Nantes, Natural Language Processing Team [1]
Starting date: Jan-2024 (flexible) ~6 months
Supervisors: Christine Jacquin and Nicolas Hernandez

Description:

Legal documents (contracts, case law, etc.) are often long and use specialised language that makes them tedious to read and difficult to understand.

The aim of this project is to study the automatic generation of section headings for US legislation which, although logically structured in paragraphs and sections, often consists of several hundred pages and no headings.

Various approaches may be considered: Keyword extraction/generation techniques, Single/Multi-documents based methods, Large Language Model, Generative models.

For the purposes of comparison and transfer of techniques, the work will be carried out on data from both the European Court of Human Rights (whose case law structure has titles) and the Supreme Court of the United States.

This research is funded by National French Research Agency under the Lexhnology project ANR-22-CE38-0004 [2]

[1] http://taln.ls2n.fr
[2] https://lexhnology.hypotheses.org/

Application:

We are looking for applications from students preparing a Master’s degree or equivalent with solid skills (and ideally experience) in Natural Language Processing, Machine Learning and Deep Learning. Excellent verbal and written skills (in French and English) are also essential.

To apply for this position, please send an email with your Curriculum Vitae, a document with your academic results, and a few words explaining your interest in this project to Christine Jacquin AND Nicolas Hernandez (firstname.lastname@univ-nantes.fr).