PhD Researcher

At the University of Göttingen -Public Law Foundation-, Institut für Informatik, there is a position as

PhD Researcher - Paraphrase Types & LLM (all genders welcome)

Entgeltgruppe 13 TV-L/100%

to be filled. Starting date is 2/1/2026. The position is limited to

The Chair of Scientific Information Analytics, headed by Prof. Dr. Bela Gipp (GippLab — ), conducts research at the intersection of computer science, data science, and information science. Natural language processing (NLP) based on Large Language Models (LLM) is a particular research focus of the chair.

As part of the research project Paraphrase Types: A New Paradigm for Paraphrase Generation and Detection (Paraphrase Types), we are looking for 1 full-time position for a PhD Researcher (m/f/d) to start by for 36 months).

Short Project Description: Paraphrases are texts conveying the same meaning using different words or grammatical structures. Humans naturally understand that changing the grammatical structure or even a word, e.g., a negation, can change the meaning of a sentence completely. Current automated systems for paraphrase generation and detection (PGD) produce and identify semantically similar content reliably. However, they only perform binary assessments of whether sentence pairs share the same meaning and fail to understand the linguistic characteristics and syntactic or semantic changes that make two texts alike. Defining and recognizing paraphrase types, i.e., different linguistic forms of paraphrases, allows us to understand what changes make two texts similar. A technique that generates and identifies paraphrase types would open many use cases. It could identify and differentiate authors more granularly, create linguistic profiles of authors, or characterize machine-generated text to improve plagiarism detection systems. Further, this technology could enhance language learning platforms, e.g., by providing learners with personalized variations of structures they struggle with, such as practice sentences with different modal verbs for those struggling with this verb type. By not considering paraphrase types in their architectures, current methods struggle with these tasks. This project will design, implement, and evaluate an approach to learn paraphrase types in large language models (LLMs) by completing three research tasks. We will assess the handling of paraphrase types in paraphrase models (WP1), integrate paraphrase types into training objectives and datasets (WP2), and develop a PGD system that incorporates these insights (WP3). In WP1, we will propose a unified taxonomy for paraphrase types and explore current LLMs limitations in PGD. In WP2, we will conduct human studies to assess paraphrase types and propose tasks and datasets for training automated systems. Next, we will formulate training tasks and compose datasets for training new models. We will propose a new metric that considers paraphrase types to evaluate models' abilities to handle specific linguistic changes. In WP3, we will implement specific LLMs for generation and detection. We will use the newly created datasets to test architecture variations and scale the best-performing models. To ensure the project's long-term success, we will develop strategies for incorporating new paraphrase types and enhancing our models with efficient computational methods. All project outputs will be made available and maintained as open-source on GitHub to ensure long-term accessibility for further research and development.

Areas of Responsibility:

• Create the first unified taxonomy of paraphrase techniques for both English and German, defining the building blocks of textual similarity.
• Investigate what cutting-edge LLMs already know about paraphrasing and expose their hidden limitations.
• Develop novel tasks, datasets, and benchmarks to train specialized language models that understand and utilize paraphrase types.
• Explore interpretability mechanisms to understand the behaviour of LLMs
• Develop new metrics for PGD
• Expand and apply the developed techniques in other projects in our group.
• Transform your research into a viable prototype—an interactive web application that can generate and detect paraphrases with surgical precision.
• Publish your findings at leading NLP conferences (e.g., ACL, EMNLP, NAACL).
• Teaching in accordance with the PhD program in Computer Science at the University of Göttingen (low workload).
• Supervise Master's and Bachelor's students working on related projects.

Your Profile:

• A master's degree (or equivalent) in computer science, computational linguistics, data science, or a related discipline. Or about to obtain a master's degree in any of the mentioned areas.
• Strong programming skills, particularly in Python.
• Hands-on experience with deep learning frameworks (e.g., PyTorch) and NLP libraries (e.g., Hugging Face).
• Solid foundation in NLP downstream tasks (e.g., NER, tokenization, word sense disambiguation).
• Professional fluency (C1) in German and English, both written and spoken.
• Communicative and teamwork-oriented mindset.

We Offer:

• A stimulating, interdisciplinary, and international research environment at one of Germany's top-tier universities.
• The chance to contribute to a high-impact project at the intersection of AI, linguistics, and cultural heritage.
• Close, supportive supervision to foster both your academic and personal development toward a PhD.
• Access to cutting-edge HPC resources, such as a GPU cluster with NVIDIA A100 GPUs, essential for large-scale model training.
• Funding to present your research at premier international conferences and engage with the global scientific community.
• A full-time position with a competitive salary under the German public service scale (TV-L), with the option of part-time employment.
• Office space and technical equipment. The position will be located in Göttingen on-site.

This role requires regular on-site presence in Göttingen; remote work is possible to a limited extent.

For more information, please contact PD Dr. Terry Ruas via email: -

The University of Göttingen is an equal opportunities employer and places particular emphasis on fostering career opportunities for women. Qualified women are therefore strongly encouraged to apply in fields in which they are underrepresented. The university has committed itself to being a family-friendly institution and supports their employees in balancing work and family life. The University is particularly committed to the professional participation of severely disabled employees and therefore welcomes applications from severely disabled people. In the case of equal qualifications, applications from people with severe disabilities will be given preference. A disability or equality is to be included in the application in order to protect the interests of the applicant.

Please upload your application in one pdf file including the usual documents until 11/26/2025 on the application portal of the university using this link: . For more information get in touch with PD Dr. Terry Lima Ruas directly via E-Mail: -, Tel

Please note:

With submission of your application, you accept the processing of your applicant data in terms of data-protection law. Further information on the legal basis and data usage is provided in the Information General Data Protection Regulation (GDPR)