Simon Ostermann

Background

About Simon Ostermann

I am a senior researcher and deputy director at the Multilinguality and Language Technology (MLT) lab at the German Research Center for Artificial Intelligence, where I am currently leading mid-sized research projects. I lead the research group on Efficient and Explainable NLP (E&E) at the MLT lab. My research broadly focuses on making language technology more accessible and fair. I focus on transparent and robust language models and language technologies. Specifically, I aim to make the parameters and behavior of language models more interpretable to both end users and researchers, drawing on methods from Explainable Artificial Intelligence and Mechanistic Interpretability. I also work on improving the robustness and efficiency of language models, particularly by enhancing their data efficiency using structured inputs, novel learning and adaptation techniques, and multimodal approaches. I especially focus on Low-Resource Languages. Additionally, I am interested in reducing model size to enable deployment in resource-constrained environments. In side projects, I'm also exploring the factuality of model-generated outputs and I have worked on multimodal models. My work primarily involves large language models trained using deep learning techniques. I also have extensive experience in data acquisition and crowdsourcing, as well as in educational NLP.