Nokia Bell Labs Internship Reliable Robust LLM Applications PhD 

Belgium 

480975613

flexible, to be agreed (typically 3-4 months), starting time is flexible

In this PhD internship, you will do research to help increase the trustworthiness of LLMs for industry applications. Potential research questions could be:

• How can we better measure the uncertainty/confidence LLMs have in their predictions? #callibration #uncertaintyestimation
• How can we better guard against LLMs leaking sensitive information? #dataprivacy
• What algorithms can be used for linking LLM predictions to the training data that influenced these predictions? #dataprovenance

Tasks

• Perform a literature review
• Design & implement algorithms / a system that help improve the reliability of GenAI applications (see the potential research questions above).
• Evaluate the effectiveness of your approach on representative problems.
• Consolidate your research in a scientific publication.
  

• Student enrolled inPh.D. Computer Science/Engineeringin Machine Learning
• Strong programming skills in Python
• Language skills: English
• Experience in AI w.r.t. safety, uncertainty quantification, robustness, interpretability, or privacy is a big plus.
• A strong publication record is a big plus.

Nokia has received the following recognitions for its commitment to inclusion & equality:

• One of the World’s Most Ethical Companies by Ethisphere
• Gender-Equality Index by Bloomberg
• Workplace Pride Global Benchmark