Summarization of German Court Rulings

Historically speaking, the German legal lan- guage is widely neglected in NLP research, especially in summarization systems, as most of them are based on English newspaper ar- ticles. In this paper, we propose the task of automatic summarization of German court rul- ings. Due to their complexity and length, it is of critical importance that legal practition- ers can quickly identify the content of a ver- dict and thus be able to decide on the rele- vance for a given legal case. To tackle this problem, we introduce a new dataset consist- ing of 100k German judgments with short sum- maries. Our dataset has the highest compres- sion ratio among the most common summa- rization datasets. German court rulings con- tain much structural information, so we cre- ate a pre-processing pipeline tailored explic- itly to the German legal domain. Additionally, we implement multiple extractive as well as abstractive summarization systems and build a wide variety of baseline models. Our best model achieves a ROUGE-1 score of 30.50. Therefore with this work, we are laying the crucial groundwork for further research on German summarization systems.