Link: https://jobs.helsinki.fi/job-invite/5299/

The positions are for a postdoctoral researcher to join a multi-disciplinary research team funded by Academy of Finland / Research Council of Finland, and the prestigious Strategic Research Council. Roman Yangarber leads the project, with additional support from experts in related disciplines.

*Position Title: Postdoctoral researcher*

*Project*:

* Explainable & Trustworthy AI for Global Threat Analysis (“Know-AI”) * AI for language teaching and language learning (“DALAI-FIN”)

*University of Helsinki – Department of Digital Humanities, Language Technology*

*Position type*: Full-time postdoctoral researcher *Duration*: initially 18 or 24 months (with possibility of extension) *Start date*: immediate

*Overview of Projects*:

* The Know‑AI Project investigates what language models know and when they know it.  The mission is to develop explainable and trustworthy AI systems for detecting and analyzing global threats, including disinformation and propaganda in online media. The successful candidate will contribute to cutting-edge research in large language models — their interpretability, robustness, and real‑world safety. * The DALAI-FIN project focuses on supporting second-language teaching for adult learners, for example, for migrants in Finland (or elsewhere in Europe), who require linguistic skills for professional and social integration.  Learners have different levels of preparation and different backgrounds. Research goals include the development of AI-based tools for personalized teaching, i.e., exactly suitable for the current level of competency of each individual learner.  This implies detailed and accurate learner modeling, learning analytics and assessment of the learner’s skills, on a continual basis.

The projects cooperate with an international network of collaborators.

*Position Description*

As a postdoctoral researcher in these projects, you will:

* Conduct foundational and applied research on explainability and trust in language models. * Develop methods to evaluate, interpret, and improve how LLMs detect and represent potential threats or risk scenarios. * Develop novel methods for learner analytics, language technology to support language teaching, assessment of difficulty, simplification, modeling the learning domain. * Collaborate closely with project research teams and external partners, including interdisciplinary groups. * Implement algorithms and data analysis (e.g., using PyTorch, JAX, transformers). * Publish papers at premier AI/ML/NLP venues and domain-related conferences. * Present results at academic conferences and workshops.

*Requirements*

* A PhD degree (or imminent completion expected) in Natural Language Processing, Machine Learning, Computer Science, or related discipline. * Experience with and strong interest in NLP, large-language models, explainability. * Demonstrated programming skills in Python, ideally including frameworks, such as PyTorch and HuggingFace Transformers. * Prior peer-reviewed publications (required for postdoctoral appointment). * Excellent oral and written communication in English.

*Salary & Benefits*

* *Starting salary:* ~3 7700–4 400 EUR/month before taxes, consistent with salary scales of the University of Helsinki. Researchersjobknow-ai-uh.github.iohelsinki.fi+1EURAXESSJobly+5helsinki.fi+5EURAXESS+5. * Comprehensive employee benefits including occupational health care, health insurance, sports facilities, and professional development support. * In Finnish universities, the teaching and research staff (app. 65% of total staff) have a special system regarding the free time equivalent to holiday. They belong to the 1,612 annual working hour system, in which the annual amount of paid holiday is app. 5-6 weeks on average per year.

*How to Apply*

Please submit via the application form a PDF file that contains:

* Curriculum Vitae, including publications and relevant projects. * A cover letter (1–2 pages) detailing your motivation, research interests, and fit with the projects. * Copies of degree transcripts. * Contact for two references.

Applications will be accepted until the positions are filled.

In case of questions, contact roman.yangarber@helsinki.fi

****In case you are at _EMNLP in Budapest_, contact now for a possible in-person meeting****