SweClinEval
Dnr:
sens2025042
Type:
NAISS SENS
Principal Investigator:
Hercules Dalianis
Affiliation:
Stockholms universitet
Start Date:
2025-09-03
End Date:
2026-10-01
Primary Classification:
10208: Natural Language Processing
Webpage:
Allocation
- Cygnus /proj at UPPMAX: 1000 GiB
- Cygnus /proj/nobackup at UPPMAX: 128 GiB
- Bianca at UPPMAX: 10 x 1000 core-h/month
Abstract
In this project generative large language models are going to be used to evaluate the performance of several downstream clinical tasks for de-identified and pseudonymised Swedish electronic patient records.