```html
Are you eager to work on a combination of Large Language Models (LLMs) with Knowledge Graphs (KGs) to create trustworthy conversational AI? Do you want to have an impact on the world’s supplier to the semiconductor industry (ASML)?
Position: PhD student
Irène Curie Fellowship: No
Department(s): Mathematics and Computer Science
FTE: 1.0
Date off: 29/09/2024
Reference number: V32.7705
This is a 4-year paid PhD position. The position will be with the Data and AI cluster at the Eindhoven University of Technology (TU/e) and ASML:
You will be supervised by Dr. J.M. Tomczak (TU/e), Prof. M. Pechenizkiy (TU/e), Prof. G. Fletcher (TU/e), and Dr. J. Kustra (ASML). You will be working in close collaboration with the Diagnostics & Data Science Group in ASML Research. This multidisciplinary team focuses on fundamentally exploring and prototyping the next generation knowledge-informed solutions for ASML, Metrology and Lithography challenges. Given the system complexity, a core challenge is in the diagnostics of (rarely occurring) failures, where the existing knowledge on system design is brought together with physics understanding as well as system data to reason on the problem potential root causes. You will participate in cutting-edge research, publish your work in leading conferences (NeurIPS, ICML, ICLR, AISTATS, UAI) and journals (TML, IEEE TPAMI, JMLR), and contribute to open-source tools.
You will work on developing a framework that will assist engineers in their diagnostics work and, consequently, shorten the downtime of a system. Additionally, the following assumptions are considered: (i) the framework must be conversational, i.e., an engineer must be able to check facts and procedures quickly, (ii) the framework must be trustworthy, namely, it cannot “hallucinate”.
We propose to formulate KG-enhanced LLMs that could serve for training, inference, and interpretability. LLMs are well-known for knowledge acquisition from large-scale systems and for achieving state-of-the-art performance on many natural language processing tasks. However, they can suffer from various issues, such as hallucinations, false references, and made-up facts. On the other hand, KGs can store enormous amounts of facts in a structured and explicit manner. However, unlike LLMs, formulating KGs is a laborious process, and querying KGs might be computationally demanding. One interesting research question is then the following: How to combine KGs and LLMs such that LLMs provide answers based on facts and do not hallucinate in any way? This could serve as a starting point for this Ph.D. project.
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:
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.
Curious to hear more about what it’s like as a PhD candidate at TU/e? Please view the video.
Information
Are you inspired and would like to know more about working at TU/e? Please visit our career page.
Do you recognize yourself in this profile and would you like to know more? Visit our website for more information about the application process or the conditions of employment. You can also contact Dr. J.M. Tomczak (j.m.tomczak@tue.nl), Prof. M. Pechenizkiy (m.pechenizkiy@tue.nl), Prof. G. Fletcher (g.fletcher@tue.nl), and Dr. J. Kustra (jacek.kustra@asml.com)
Application
We invite you to submit a complete application by using the apply button. The application should include a:
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.