Benedikt Ebing

Benedikt Ebing is a Research Associate and Doctoral Student at the CAIDAS Chair for Natural Language Processing at the University of Würzburg, Germany. He is based in Building 50 (Campus Nord) and focuses on translation-based cross-lingual transfer, multilingual NLP, and dataset development for political and low-resource language scenarios.

• Current affiliation: University of Würzburg (CAIDAS Chair for NLP)
• Research areas: NLP, cross-lingual transfer, low-resource languages, dataset curation
• Key contributions: ParlEE dataset, Kardeş-NLU benchmark, analysis of translation-based transfer
• Email: benedikt.ebing@uni-wuerzburg.de

His recent work explores token classification across languages, Turkic language benchmarks, and ethical considerations in translation-based transfer models. The ParlEE dataset provides extensive annotated speech data for EU states, enabling large-scale political text analysis.