Applications are invited for a fully funded 4-year PhD Position in Post-Training of Diffusion Language Models within the Rhine AI research group led by Prof. I. Bogunovic at the Department of Mathematics and Computer Science, University of Basel. This doctoral opportunity is funded by the Swiss National Science Foundation (SNSF) under the project “Foundations of Post-Training for Reasoning in Diffusion Large Language Models.”
Overview
| Program / Position | PhD Position (100%) |
|---|---|
| Institution | University of Basel |
| Department / Group | Department of Mathematics and Computer Science / Rhine AI |
| Research Supervisor | Prof. I. Bogunovic |
| Research Area | Machine Learning, Artificial Intelligence, Diffusion Language Models (dLLMs), Reinforcement Learning |
| Location | Basel, Switzerland |
| Duration | 4 Years (Full-time) |
| Start Date | January 2027 or later |
| Preferred Deadline | November 20, 2026 (Open until filled) |
Designation
PhD Researcher / Doctoral Candidate (100%)
Research Area
Machine Learning, Deep Learning, Generative Modeling, Post-Training of Diffusion-Based Language Models (dLLMs), and Reinforcement Learning for Advanced Reasoning.
Scholarship Description
This doctoral project focuses on the post-training of large diffusion-based language models (dLLMs), which represent a prominent and emerging alternative to standard autoregressive LLMs, specifically aiming to enhance their reasoning capabilities. Key responsibilities and opportunities include:
- Developing reinforcement learning methods designed for post-training large dLLMs.
- Investigating reasoning-aware generation approaches, such as adaptive unmasking and credit assignment.
- Constructing benchmarks to assess reasoning performance across complex domains, including mathematical problem-solving.
- Conducting large-scale empirical experiments backed by principled theoretical analysis.
- Publishing scientific contributions in premier machine learning conferences such as ICML, NeurIPS, and ICLR.
- Collaborating closely with leading national and international AI teams, including researchers at ETH Zurich and EPFL.
- Receiving competitive Swiss doctoral salaries, comprehensive benefits, and conference travel support.
Location
Basel, Switzerland.
Eligibility / Qualification
Candidates must satisfy the following criteria:
- A completed Master’s degree in Computer Science, Mathematics, Physics, or a closely related quantitative field.
- An outstanding academic record with strong foundational knowledge in mathematics.
- Proficiency in programming and implementation, with hands-on capability to conduct large-scale machine learning experiments.
- Background or prior exposure to LLMs, reinforcement learning, or generative modeling is considered an advantage.
- Prior scientific research experience is beneficial but not strictly mandatory.
- Fluency in written and spoken English, paired with excellent communication skills.
- Women and individuals from underrepresented backgrounds in computer science and mathematics are strongly encouraged to apply.
How to Apply
Interested candidates should prepare their application in two separate PDF files covering the following materials:
- Document 1: A brief motivation statement (maximum 150 words) outlining your interest in the position and why your academic background makes you a strong fit.
- Document 2: A comprehensive dossier containing:
- Curriculum Vitae (including a list of publications, if applicable).
- Scanned copies of academic transcripts (Bachelor’s and Master’s degrees).
- Contact details of academic or professional references (do not send letters directly).
- Your Master’s thesis and/or representative publications, if available.
Last Date for Apply
Candidates are strongly encouraged to submit their applications before November 20, 2026. Applications will continue to be reviewed on a rolling basis until the position is filled.
Apply Link
To submit your application or learn more details, please visit the official job portal:
Apply for the PhD Position at University of Basel
