PhD Fellowship(s) in AI Safety and Mechanistic Interpretability of Language Models
Share
Job Location:
Monthly Salary:
Posted on:
30+ days ago
Vacancies:
1 Vacancy
Job Summary
The Centre for Machine Learningwithin the Data Science and Statistics Sectionof the Department of Mathematics and Computer Science (IMADA)at the University of Southern Denmark invites applications for PhD research fellowship position(s) within the field of machine learning AI Safety and mechanistic interpretability to be filled earliest by 1 January 2026 for a period of three years.
The research will be conducted within the MIST project (Scalable Mechanistic Interpretability for Safe and Trustworthy LLM Agents) recently funded by the Novo Nordisk Foundation. The project aims to develop scalable methods for understanding the inner workings of large language models (LLMs) and LLM agents with a focus on identifying causal mechanisms underlying tool use reasoning and multi-agent communication. The research will investigate cross-model and cross-lingual universality of these mechanisms and develop functionally grounded steering techniques and methods to issue safety certificates.
Research topics include (but are not limited to) interpretability and transparency agentic and multi-agent safety control and containment safety evaluation sustainability and resource impact. Selected candidates may also contribute to ongoing work in the Danish Foundation Models project particularly on multilinguality evaluation and efficiency aspects of language models.
We are seeking candidates with:
- Strong desire to make significant contributions to science and specific interest in AI Safety and mechanistic interpretability research
- Creative and independent thinking to develop novel approaches to challenging research problems
- Outstanding theoretical background in machine learning and deep learning demonstrated by excellent course grades and where applicable by research experience or scientific publications
- Strong programming skills in Python and deep learning frameworks such as PyTorch demonstrated by coursework projects or contributions to public code repositories
- Experience with or strong interest in large language models transformers and natural language processing
- Interest in multilingual models is a plus
- Excellent spoken and written communication skills in English
- Background in interpretability methods causal inference or multi-agent systems is advantageous but not required
The successful candidate will have the unique opportunity to contribute to establishing a new research group on AI Safety at SDU and will participate in publishing high-quality research papers at top-tier machine learning and NLP venues such as NeurIPS ICLR ACL and EMNLP. The candidate will also fulfill teaching assistantship duties.
We will consider candidates who have (or will obtain before the start date) a Masters degree in Computer Science Data Science Artificial Intelligence Mathematics Statistics or related fields. Candidates should demonstrate that they have passed at least two Masters level courses that cover advanced machine learning topics with a grade that corresponds to the top 10% within the native grade scale or can show comparable achievements such as a Masters project focusing on topics relevant to the PhD position.
IMADA has the unique feature of bringing mathematicians and computer scientists together within a single department to foster theoretically well-backed high-quality data science research. IMADA is home to many ongoing externally funded research projects as well as to a rich curriculum of data science and artificial intelligence courses. Data Science and Statistics Group is a synergy platform for the data science experts in IMADA.
Place of work
The Department of Mathematics and Computer Science is located at the main campus of the University of Southern Denmark Odense Denmark. The University of Southern Denmark was founded in 1966 and now has more than 27000 students almost 20% of whom are from abroad. It has more than 3800 employees and 115 different study programmes in the fields of the humanities social sciences natural sciences health sciences and engineering. Its main campus is located in Odense the third largest city in Denmark.
Odense provides family-friendly living conditions with the perfect combination of a historic city centre with an urban feel and yet a close proximity to beaches and recreational areas. Its location on the beautiful island of Funen is ideal with easy access by train or highway to the bigger cities of Aarhus and Copenhagen. As the birthplace of Hans Christian Andersen Denmarks famous fairytale author the city is home to a vibrant and creative population that hosts numerous festivals and markets throughout the year.
Contact information
For further questions about the position please contact Assistant Professor Lukas Galke Poech at .
If you experience technical problems please
Application salary etc.
Appointment as a PhD Research Fellow can be in different types of programmes
The 53 programme: Appointment as a PhD Research Fellow is for three years. Students who have completed their MSc. Employment stops automatically at the end of the period.
Further information about the PhD programme at the Faculty of Science can be found at thehomepage of the University.
The University wishes our staff to reflect the diversity of society and thus welcomes applications from all qualified candidates regardless of personal background.
Application must be in English and made in the form of a Declaration of Interest including the following:
- A letter stating your specific interest motivation and qualifications for the project in question (max. two pages) .
- Detailed CV including personal contact information anddate entries
- Copies of diplomas Bachelor as well as Masters degree including transcript of notes/grades.
- The names and complete contact details (phone numbers and e-mail) of at least two referees details for at least 2 referees who can vouch for your skills.
Shortlisting may be used in the assessment process.Incomplete applications and applications received after the deadline will neither be considered nor also applies to reference letters.
Applications will be assessed by an expert assessor/committee. Applicants will be informed of their assessment by the university.
Applications should be sent electronically via the link Apply now. The faculty expects applicants to read the information How to apply for a position at SDU before applying. Please note that the application documents should be in pdf-format and attached as three documents: motivation letter CV and the remaining documents.
We recommend that as an international applicant you take the time to visitWork in Denmarkwhere you will find information and facts about moving to working and living in Denmark as well asthe International Staff Office at SDU.
Key Skills
- Security Management
- Vehicle Patrols
- Public Safety
- Law Enforcement
- Access Control
- Safety Procedures
- Security Measures
- Alarm System
- Crowd Control
- CCTV Monitoring
- Access Point
- Security Checks
- Detect Signs
- Safe Environment
- Security System
About Company
The University of Southern Denmark was established to create value for and with society. Whether our contributions come in the form of excellent research, innovative solutions, education or learning, we must make a positive difference to society and contribute to a sustainable future. ... View more
