PhD in Tensor Factorization Models for Creative Language Generation

About the role

The Centre for Computational Linguistics at KU Leuven is looking for a PhD researcher to advance tensor factorization models for creative language generation within the ERC‑funded TENACITY project. The work will explore how latent semantic dimensions derived from multi‑way word co‑occurrences can be interpreted and manipulated to produce novel linguistic expressions.

Key responsibilities

• Develop and evaluate efficient tensor factorization algorithms for large‑scale, sparse language data.
• Design methods to automatically induce semantic domain mappings.
• Apply latent semantic structures to a range of creative language phenomena, extending current work on metaphor.
• Collaborate with a team of PhD researchers and post‑docs, contributing to publications and conference presentations.

Required profile

• Master’s degree in computational linguistics, AI, computer science, mathematics or a related field.
• Strong background in machine learning and linear algebra.
• Interest in lexical semantics, figurative language and computational creativity.
• Excellent proficiency in English and good communication skills.

Required skills

• Python programming (NumPy/SciPy ecosystem).
• Machine learning techniques.
• Linear algebra concepts.
• Familiarity with tensor methods (a plus).

What we offer

• Full‑time PhD position (1 year, extendable to 4 years based on performance).
• Access to high‑performance computing resources and conference travel funding.
• Opportunity to work in a prestigious ERC‑funded research environment.
• Academic training, workshops, and a pathway to a doctoral degree.