PhD-Position - Novelty Assessment of Science and Patents with LLMs

Place of Employment: Jülich
Start Date: To the next possible date
Salary: Pay group 13 (75%) TVöD-Bund

Application Deadline: The position will be advertised until it is successfully filled.

Index number: 2026D-0626

Your Job

Background information about the project: Our team recently won the Metascience Novelty Indicator Challenge. We have been developing a method that uses Large Language Models (LLMs) to analyze the content of scientific articles, alongside relevant prior work. The method assesses the novelty of contributions by reasoning about their content, rather than counting citations.

You will further develop the method and apply it systematically to scientific literature and patents, focusing on technologies relevant to analyzing energy transition pathways. This will provide valuable insight into the directions in which technology and research fields are advancing.

Furthermore, you will also extend the indicator beyond novelty toward quality assessment and validate the reliability, robustness, and fairness of LLM-based research indicators. The main use case for assessing the quality of publications is to improve the reliability of literature-derived data in our technology database. Ensuring the quality of this database is essential because energy system models and their results depend on it.

Why It Matters: Climate and energy systems research require an up-to-date understanding of the state of research across many technologies to inform modeling assumptions. Better data on the novelty and quality of scientific contributions strengthens the research that informs decarbonization and is highly relevant to the metascience community in general. Additionally, there is a high demand from researchers, publishers, and funders for LLM-based research indicators to help them identify relevant publications amid the surge in AI-generated content.

Your Tasks in Detail:

• Advance LLM-based methods for automatic novelty assessment of patents and scientific literature
• Map the state of the art of scientific fields
• Investigate and evaluate the approach along the dimensions of reliability, robustness, transparency, and fairness, and characterize its opportunities and limitations
• Extend the indicator to further notions of research quality, using large-scale, literature-derived, technology-parameter databases as the driving use case
• Apply the method to improve energy system analysis in collaboration with energy system modelers

Your Profile

• Completed master's degree in a STEM field; a degree in computer science, data science, or an AI-related discipline is preferred
• Strong interest in analyzing scientific literature and patents
• Strong interest in energy technologies and energy systems research
• Experience in LLMs and developing LLM-based methods is an advantage
• Programming skills, ideally in Python
• Structured, analytical, and independent working style
• Reliability, attention to detail, and enjoyment of working in an interdisciplinary team
• Fluent English, both written and spoken (C1/C2 level according to the CEFR), ideally supported by a certificate confirming the language level); German language skills are an advantage

Please feel free to apply even if you do not yet meet all the required skills and qualifications. We may be able to provide training for these during the onboarding process.

Our Benefits for You

We work on the very latest issues that impact our society and are offering you the chance to actively help in the change! We support you in your work through:

• Team & Environment: A highly motivated, international, and interdisciplinary team within one of Europe’s largest research institutions
• Collaborative Research: Participation in exciting collaborative research projects with top-tier partners
• Professional Development: Excellent further training, cross-disciplinary collaboration, and practical insights into interdisciplinary challenges at the intersection of research data management, data science, and artificial intelligence
• Excellent Infrastructure: Outstanding scientific and technical infrastructure
• Scientific Mentoring: Continuous professional guidance from your scientific supervisor
• Remote work: Ideal conditions for successful remote work (after consultation and in line with upcoming tasks and (on-site) appointments)
• Work-Life Balance: Flexible working hours and 30 days of vacation
• Personal Development: Development of your personal strengths, e.g., through our extensive training program and networking opportunities specifically for doctoral candidates (JuDocS): https://go.fzj.de/JuDocs
• Fair Remuneration: Depending on your qualifications and assigned responsibilities, you will be classified according to pay group 13 (75%) of the TVöD-Bund. Additionally, you will receive a special payment (“Christmas bonus”) amounting to 75% of one month’s salary. All information about the TVöD-Bund collective agreement can be found on the BMI website (pay scale table on page 69 and following of the PDF download): https://go.fzj.de/bmi.tvoed
• Contract: The employment of doctoral researchers at Jülich is governed by a doctoral contract, which usually has a term of three years.
• Support for International Employees: Our International Advisory Service makes it easier for international employees to get started

Please feel free to apply either in English or in German.

In addition to exciting tasks and a collegial working environment, we offer much more: https://go.fzj.de/Benefits

We welcome applications from people with diverse backgrounds, e.g. in terms of age, gender, disability, sexual orientation / identity, and social, ethnic and religious origin. A diverse and inclusive working environment with equal opportunities in which everyone can realize their potential is important to us.

The following links provide further information on diversity and equal opportunities: https://go.fzj.de/equality and on specific support options for women: https://go.fzj.de/womens-job-journey