PhD on Philosophical Constraints on the Explainability and Fairness of LLMs

Position PhD-student

Irène Curie Fellowship No

Department(s) Industrial Engineering and Innovation Sciences

FTE 1,0

Date off 01/12/2024

Reference number V39.7844

Job description

Do you want to conduct philosophical research on the explainability and fairness of state-of-the-art large language models? Do you have an interdisciplinary background that spans philosophy, artificial intelligence, and cognitive science? Are you interested in conducting impactful philosophical research in close collaboration with scientists and engineers in a EU-funded doctoral network, helping to promote fair and transparent generative AI?

We are seeking a highly motivated doctoral candidate to join our research team as part of a prestigious Marie Skłodowska-Curie Actions (MSCA) doctoral network: AlignAI. The AlignAI doctoral network aims to train doctoral candidates to develop, evaluate, and engage with Large Language Models (LLMs). It focuses on aligning these models with human values to ensure their development and deployment are ethically sound and socially beneficial. By integrating expertise from social sciences, humanities, and technical disciplines, the project will address critical issues such as explainability and fairness, thereby ensuring LLMs contribute positively to education, mental health, and news consumption.

The doctoral candidate in this position will develop a philosophical framework to ensure and evaluate the explainability and fairness of Large Language Models. This framework will center on conceptual and theoretical tools with which to understand, systematize, and evaluate formal and computational methods from AI that help explain the behavior of state-of-the-art LLMs, and that help align such systems with human values generally and fairness-related values specifically. Depending on the interest and expertise of the candidate, these tools might be designed to attribute conceptual or cognitive capacities to artificial neural networks on the basis of their learned representations, to apply benchmarks or statistical measures to evaluate the fairness and toxicity of their generated content, and/or to evaluate the success of probing and intervention techniques to manipulate their behaviors.

The doctoral candidate will need to apply methods from philosophy of science, philosophy of mind, and ethics to recent work in artificial intelligence, psychology, neuroscience, linguistics, and/or related disciplines. They will also need to learn about contemporary LLM applications in education, mental health, and online news consumption. Thus, the ideal candidate will have an interdisciplinary background that spans several of the aforementioned disciplines, and that demonstrates a clear interest in AI technology and its societal impact. At the same time, the candidate will have the necessary language, communication, and analytic skills to produce philosophical research of publishable quality.

The doctoral candidate will be supervised by Carlos Zednik and Lambèr Royakkers, and will be embedded within the Philosophy & Ethics group, Department of Industrial Engineering and Innovation Sciences, Eindhoven University of Technology (TU/e). Additional supervision will be provided within the AlignAI doctoral network by Andrea Cavallaro (EPFL) and Christoph Lütge (TU München). The candidate will be affiliated with the Eindhoven Center for Philosophy of AI (ECPAI) and the Eindhoven Artificial Intelligent Systems Institute (EAISI), and will be expected to contribute actively and positively to the philosophical and AI research communities in Eindhoven. The candidate will also be expected to attend workshops, summer schools, and research visits at AlignAI partner institutions.

This position is one of four PhD positions being offered at TU/e through the AlignAI doctoral network, the others being in Human Technology Interaction (Martijn Willemsen) and Industrial Design (Jesse Benjamin & Stephan Wensween).

Job requirements

• Candidates must have a bachelor and/or master’s degree in philosophy, and additional formal training in computer science, cognitive science, psychology, neuroscience, linguistics, or a closely related discipline.
• Applicants must not own a doctoral degree at the date of recruitment.
• Applicants may be of any nationality, but must not have resided or carried out their main activity (work, studies, etc.) in the Netherlands for more than 12 months in the 36 months immediately before the recruitment date.
• Candidates must be able to work in an interdisciplinary team and interested in collaborating with industrial partners.
• Candidates must be motivated to develop their teaching skills and coach students.
• Candidates must be fluent in spoken and written English (C1 level).

Selection Procedure

• Application Submission: Applications must be submitted by December 1, 2024.
• Selection Criteria: Candidates will be evaluated based on their academic background, research experience, and alignment with the project’s objectives. The selection process will be open, transparent, merit-based, impartial, and equitable.
• Interview: Shortlisted candidates will be invited for a structured interview with predefined questions and a scoring system.
• Reserve List: A reserve list of top candidates may be created for future opportunities.
• Selection Committee: The selection committee will comprise diverse experts ensuring gender balance, nationality diversity, and relevant experience.

Conditions of employment

A meaningful job in a dynamic and ambitious university, in an interdisciplinary setting and within an international network. You will work on a beautiful, green campus within walking distance of the central train station. In addition, we offer you:

• Full-time employment for four years, with an intermediate evaluation (go/no-go) after nine months. You will spend 10% of your employment on teaching tasks.
• Salary and benefits (such as a pension scheme, paid pregnancy and maternity leave, partially paid parental leave) in accordance with the Collective Labour Agreement for Dutch Universities, scale P (min. €2,872 max. €3,670).
• A year-end bonus of 8.3% and annual vacation pay of 8%.
• High-quality training programs and other support to grow into a self-aware, autonomous scientific researcher. At TU/e we challenge you to take charge of your own learning process.
• An excellent technical infrastructure, on-campus children's day care and sports facilities.
• An allowance for commuting, working from home and internet costs.
• A Staff Immigration Team and a tax compensation scheme (the 30% facility) for international candidates

Information and application

About us

Eindhoven University of Technology is an internationally top-ranking university in the Netherlands that combines scientific curiosity with a hands-on attitude. Our spirit of collaboration translates into an open culture and a top-five position in collaborating with advanced industries. Fundamental knowledge enables us to design solutions for the highly complex problems of today and tomorrow. 

Information

Do you recognize yourself in this profile and would you like to know more?
Please contact the hiring manager, Dr. Carlos Zednik at c.a.zednik@tue.nl.

Visit our website for more information about the application process or the conditions of employment. You can also contact hrservices.ieis@tue.nl or +31 40 247 5152.

Are you inspired and would like to know more about working at TU/e? Please visit our career page.

Application

We invite you to submit a complete application by using the apply button. The application should include a:

• Cover letter in which you describe your motivation and qualifications for the position.
• Curriculum vitae, including a list of your publications and the contact information of three references.
• Writing sample (e.g. term paper, master’s thesis, published article) which demonstrates your ability engage in philosophical reasoning about science or technology.

We look forward to receiving your application and will screen it as soon as possible. The vacancy will remain open until the position is filled.