In our paper, we assemble a broad Natural Language Understanding benchmark suite for the German language and consequently evaluate a wide array of existing German-capable models in order to create a better understanding of the current state of German LLMs.
more
Our paper "Marina: Realizing ML-driven Real-time Network Traffic Monitoring at Terabit Scale" has been accepted in the IEEE journal on Transactions on Network and Service Management. The paper is a result of our cooperation with the Prof. Dr. Tobias Hoßfeld and the chair of Communication Networks at the University of Würzburg.
more
We participated in two SemEval tasks and placed first on average in SemEval'24 Task 4!! OTTERly fantastic!
more
Our paper "CompTrails: comparing hypotheses across behavioral networks" has been published in the DAMI Journal and is open accessible.
more
At the 4th international workshop on advanced neural algorithms and theories for recommender systems we won the award for the best Paper with "Enhancing Sequential Next-Item Prediction through Modelling Non-Item Pages".
more
The MAGNET4Cardiac7T Project has been granted 40.000 GPU-hours in the Erlangen National High Performance Computing Center for training several deep learning models for simulating electromagnetic fields.
more
The dataset of 1TB has been published in our University's research repository and is available for download.
more
In this joint work with the Chair of Human-Computer Interaction (LS IX) , we introduce a new dataset 'Who is Alyx?' with behaviometric and biometric data from VR users to address a critical issue in this field, which is designed for investigating user identification based on motion in XR environments.
more