PhD Fellowship in Natural Language Processing (Cultural Reasoning with Language Models) at the University of Copenhagen
Job Description
The University of Copenhagen’s Department of Computer Science is offering a PhD fellowship in natural language processing on how language models reason about culture and interpretation, within the CuRe project funded by Independent Research Fund Denmark. Literary texts serve as a demanding testbed, and the project includes a planned research stay at McGill University.
At a glance
- Position: PhD fellowship, AI Section, Department of Computer Science
- Project: CuRe: Cultural Reasoning for Responsible Language Model Development (DFF project 5334-00088B)
- Start: 1 February 2027 or as soon as possible thereafter
- Group: CoAStaL research group
- Apply: 8 November 2026, 23:59 CET
About the NLP PhD project
- Build datasets and benchmarks that link literary passages with cultural and historical context, textual evidence and expert or reader interpretations
- Analyse what cultural knowledge and interpretive strategies language models use, where they fail, and how confidently they present uncertain conclusions
- Develop methods for grounding interpretations in evidence and representing uncertainty, ambiguity and multiple plausible readings
- Evaluate methods through automatic experiments and studies with literary scholars and other readers
- Publish in leading NLP, AI and computational humanities venues; the empirical work mainly uses modern Scandinavian literary texts
A Danish PhD combines supervised research with about 30 ECTS of PhD courses, a research stay at another group, and some teaching or dissemination, ending in a PhD thesis. Positions are advertised with a fixed project and supervisor, and applicants apply to that specific project.
What University of Copenhagen is looking for
- A strong computational background, for example in NLP, computational linguistics, computer science, machine learning, AI, data science or digital humanities
- Interest in studying language, culture and interpretation rigorously
- Motivation to work across disciplines and engage seriously with humanities research questions; formal training in literary studies is not required
Nice to have
- Experience with language model evaluation and probing, retrieval and RAG, data annotation, human-centred evaluation or multilingual analysis
Pay and location
Salary, pension and terms of employment follow the agreement between the Danish Ministry and the Danish Confederation of Professional Associations (AC); the posting gives no figure. The position is based in Copenhagen, Denmark.
How to apply
Apply through the official University of Copenhagen page. The closing date is 8 November 2026, 23:59 CET. Apply through the University of Copenhagen job portal using the “Apply now” link on the official posting, which lists the documents to include.
See all our artificial intelligence jobs, or browse more research jobs on ResearchJobs.in.
Hiring institution: University of Copenhagen
Official advertisement: employment.ku.dk
How to prepare for this application
- Evaluation and probing: read recent work on cultural knowledge and uncertainty in LLMs and be ready to critique a benchmark.
- Humanities curiosity: show genuine interest in how literary scholars interpret texts.
- Research proposal: sketch one dataset or method you would build for culturally grounded interpretation.
- Multilingual angle: the texts are mainly Scandinavian; mention any cross-lingual NLP experience.
- Code: link a repository showing LLM evaluation or RAG work.
About the University of Copenhagen
The University of Copenhagen, founded in 1479, is Denmark's oldest and largest university. In Denmark, PhD students are normally employed on full-time salaried contracts, with pay and terms set by national collective agreements, and the standard PhD lasts three years after a master's degree.
