Julian Risch

Julian Risch is a researcher at the University of Potsdam, Germany, with a focus on Natural Language Processing (NLP) and Machine Learning. His work spans topics such as aggression identification, document embeddings, comment analysis, and domain-specific text modeling. He has collaborated extensively with Ralf Krestel and other researchers across institutions.

Research Interests

• NLP: Toxic comment detection, BERT ensembles, German language processing
• Machine Learning: Pseudo-labeling, entropy-based topic modeling, multimodal retrieval
• Information Retrieval: Document similarity, patent classification, hybrid recommendation systems

Notable Articles

His recent publications highlight trends in LLM-based training data generation, domain-specific embeddings, and semantic answer similarity. He also contributed to German NLP benchmarks (GermanQuAD) and computational social science (offensive language detection).

Collaborations

Julian has partnered with leading researchers in NLP and computational linguistics, including Ralf Krestel, Alan Akbik, and Bogdan Kostic. His work addresses real-world challenges in news engagement, patent analysis, and academic knowledge graphs.