German Text Summarization Challenge

With the rise of deep learning, automatic text summarization has made promising progress. However, it is still an unsolved problem and an open research topic. We aim to explore your ideas and solutions regarding summarization of German texts. For that we invite you to participate in the 1st German Text Summarization Challenge. The results will be presented at the 4th Swiss Text Analytics Conference (SwissText 2019), which will be held June 18-19 in Winterthur, Switzerland.

With this challenge we hope to create a basis for interesting discussions and take the NLP community for German text understanding a step further.

For the challenge, we provide you with 100,000 texts together with reference summaries extracted from the German Wikipedia. The aim is to generate abstractive summarizations. In order to avoid steering the development towards certain metrics, we won’t release our metrics until the end of the challenge. The training data is already available:

Please follow this link and register:



Friday 1. March: Release of training data and submission format

Friday 31. May: Submission deadline (solution + system description)

Friday 7. June: Author Notification

18. /19. June: SwissText (announcement of ranking + presentations + discussions + prize awarding) – The submission is done by sending an E-Mail to providing the solutions to the “data_test.csv” file alongside a short system description.

We are looking forward to many interesting solutions and a lively discussion.