PhD: Constraint-Based Neural Language Models for Creative Language Generation

About the role

The Centre for Computational Linguistics at KU Leuven, within the ERC‑funded TENACITY project, is looking for a PhD researcher to investigate constraint‑based neural language models for creative language generation. The candidate will work on integrating constraints into transformer architectures to enable novel, expressive text generation.

Key responsibilities

• Design and implement methods that embed linguistic or semantic constraints into transformer‑based language models.
• Conduct experiments on creative language generation tasks and evaluate model performance.
• Collaborate with a multidisciplinary team of PhD students and post‑docs, contributing to publications and conference presentations.
• Document research findings and integrate them with the project’s latent semantic representation framework.

Required profile

• Master’s degree in artificial intelligence, computational linguistics, computer science or a related field.
• Strong background in neural networks and natural language processing.
• Good understanding of large language model and transformer architectures.
• Excellent programming abilities, preferably in Python and PyTorch.
• High proficiency in English and solid communication skills.
• Strong interest in language, linguistics and computational creativity.

Required skills

• Python
• PyTorch
• Deep learning
• Natural language processing

What we offer

• Full‑time PhD position (initially 1 year, extendable to 4 years after a positive evaluation).
• Access to high‑performance computing resources and ERC project funding.
• Support for conference travel, academic training and workshops.
• Opportunity to work in a vibrant, international research environment.