Researcher in computational linguistics focused on multimodal NLP

Reference number PAR 2026/275

The University of Gothenburg tackles society’s challenges with diverse knowledge. 58 000 students and 6800 employees make the university a large and inspiring place to work and study. Strong research and attractive study programmes attract researchers and students from around the world. With new knowledge and new perspectives, the University contributes to a better future.

The University of Gothenburg is a state authority, which means excellent employment benefits such as wellness time during working hours, compensation for medical visits, a base vacation of 28 days, and an occupational pension. You can read more about our employment benefits  here . 

At the Department of Philosophy, Linguistics, and Theory of Science (FLoV) within the Faculty of Humanities, we have around 120 employees and a teaching volume that includes around 350 full-year students. We are engaged in numerous large international and interdisciplinary research projects in areas such as theoretical and practical philosophy, logic, linguistics, computational linguistics, language technology, and theory of science. Our research has many applications and collaborations with other scientific fields, including biology, computer science, economics, history of ideas, law, medicine, psychology, religion, political science, and engineering. We educate philosophers, logicians, linguists, language technologists, theorists of science, historians of philosophy, and teachers of philosophy. The department also conducts doctoral education and development work within its various academic domains. If you are interested in working in an international, dynamic and collaborative enviornment, FLoV offers an inspiring setting for research and education. Read more about us here. 

Project description

The Centre for Linguistic Theory and Studies in Probability (CLASP) is based in FLoV at the University of Gothenburg, and is funded by a 10-year grant from the Swedish Research Council (2015-2027). CLASP is devoted to research and advanced training in the application of probabilistic modeling and machine learning methods to core issues in linguistic theory and cognition. It also addresses topics in dialogue and linguistic interaction. For a description of CLASP, see  ?url=https%3A%2F%2Fwww.gu.se%2Fen%2Fclasp&module=jobs&id=3224866" target="_blank" rel="nofollow">?url=https%3A%2F%2Fwww.gu.se%2Fen%2Fclasp&module=jobs&id=3224866" target="_blank" rel="nofollow">https://www.gu.se/en/clasp .

This project, within CLASP, investigates coherence in multimodal large vision-language models (LLMs), with a focus on how meaning is constructed across visual and textual modalities. Our central research question is how humans and LLMs establish coherence and how their underlying strategies differ. To address this, we will employ crowdsourcing methods to develop and annotate training and evaluation datasets, enabling systematic comparison between human judgments and model predictions.

In parallel, we will evaluate and fine-tune existing multimodal LLMs, using targeted benchmarks and task-specific datasets to assess their ability to model coherent cross-modal relationships and to improve their performance. The project further explores methods for integrating visual and linguistic representations into unified embedding spaces, aiming to identify the most effective approaches for supporting coherence-sensitive tasks.

Building on prior work in areas such as named entity recognition and coreference resolution, the project seeks to advance our understanding of coherence in next-generation multimodal AI systems. The candidate will carry out their research within the larger research group CLASP. 

Subject area 

Computational linguistics 

Subject area description 

The topic of coherence belongs to the broader area of discourse modelling in computational linguistics. Discourse modelling examines the mechanisms that organize sequences of sentences or utterances into meaningful, intelligible, and contextually appropriate discourse.

These mechanisms include phenomena such as coreference, explicit and implicit discourse relations, lexical consistency, and the continuity of verbal tense and aspect.

Duties

The researcher is expected to collect, compile and process corpora from different genres, domains and methodologies. They will also develop and evaluate models for multimodal NLP, particularly in the area of long-context text coherence. Furthermore, the candidate will apply these models to advance the state of the art in a range of multimodal NLP tasks, including both applications such as text generation and experimental approaches to linguistic analysis.

The holder of the position is expected to carry out their research on site at campus of the University of Gothenburg and to actively participate in workshops, seminars, and conferences. The working language of the research group is English.

Qualifications 

Eligible for a researchers position are those with a Master degree or other foreign degree that is deemed equivalent to Master. The subject of the master thesis should be in a research area relevant to the specific subject description above. 

Scientific proficiency is the most important criterion for assessment, and special attention will be given to the applicant’s Master thesis and publications if existing. Documented good interpersonal skills, good skills in terms of planning and organizing research, and good skills in communicating research results is also meritorious. 

Candidates should have experience with machine learning including deep learning and neural networks, preferably with a background in mainstream natural language processing tasks and methods (e.g., language modeling, parsing, coreference, machine translation, discourse parsing, etc). Interest in the cognitive processes/models behind human language processing and understanding is a key advantage for the position. Experience with crowdsourcing data and statistical analysis is also desirable.

Employment 

Type of employment: Fixed-term employment, 11 months
Extent: 100%
Location: Department of Philosophy, Linguistics and Theory of Science, Gothenburg
First day of employment: As soon as possible, by agreement

Salaries are set individually at the University.

Selection process

After the application deadline, selection, interviews (possibly by video link), and assessments are conducted by a recruitment group at the department.

More information regarding the University’s recruitment process you will find here.

Contact information 

For questions regarding the position, subject, research project, responsibilities, or the workplace, please contact Sharid Loáiciga, Associate Senior Lecturer, sharid.loaicigagu.se +46(0) 31-786 59 42

For questions regarding the application, recruitment process, or employment conditions, please contact HR-officer Jennifer Stråle, jennifer.stralegu.se +46 (0)31-786 4570

Unions 

Information regarding union representatives at the University of Gothenburg, see link.

Application

Submit your application via the University of Gothenburg’s vacancies website, here , select the advert for this position, and follow the link marked ’Apply’ at the bottom of the page. It is your responsibility to ensure that the application is complete as per the vacancy notice, and that the University receives it by the final application deadline.

Please note that the following documents should be attached to the application (in PDF-format):

• Personal letter of 500 words maximum (approx. 1 page). The letter must consist of three paragraphs in the following order: a) a short presentation of yourself, b) a short presentation of your merits, c) a short account for why the Department of Philosophy, Linguistics and Theory of Science and the project is of interest to you.
• A CV listing academic degrees, employments, publications, and other relevant merits.
• Master degree certificate.
• At most three scientific publications, and the applicant’s Master thesis.
• Appendices (optional)

Any documents that are not available in electronic format should be mailed to: University of Gothenburg, HR administrator, FLoV, Box 200, SE-405 30 Gothenburg, SWEDEN. Reference number should be clearly stated on the mailed documents. The documents must be received no later than three days after the application deadline, or the sender must be able to show that the documents were posted no later than the deadline for electronic applications.

We welcome applications written in English, since a non-Swedish speaking person may be a member in the recruitment process.

References for candidates in the final selection may be requested.

Applications must be received by: 2026-04-23

The University works actively to achieve a working environment with equal conditions, and values the qualities that diversity brings to its operations.

Salaries are set individually at the University.

In accordance with the National Archives of Sweden’s regulations, the University must archive application documents for two years after the appointment is filled. If you request that your documents are returned, they will be returned to you once the two years have passed. Otherwise, they will be destroyed.

In connection to this recruitment, we have already decided which recruitment channels we should use. We therefore decline further contact with vendors, recruitment and staffing companies.