EUR-Lex-Sum
A benchmark dataset for multilingual legal text summarization, covering 24 languages across EU legal documents. It is used to evaluate the performance of multilingual and cross-lingual legal summarization models.
Timeline (last 90 days)
LexLattice achieves state-of-the-art ROUGE scores on all 24 languages of the EUR-Lex-Sum dataset under both multilingual and cross-lingual settings.
EUR-Lex-Sum is a benchmark dataset constructed specifically for multilingual legal summarization research, covering all 24 official languages of the European Union, with content sourced from regulatory summaries on the EU legal database EUR-Lex.
All Facts (2)
LexLattice achieves state-of-the-art ROUGE scores on all 24 languages of the EUR-Lex-Sum dataset under both multilingual and cross-lingual settings.
50%UnverifiedEUR-Lex-Sum is a benchmark dataset constructed specifically for multilingual legal summarization research, covering all 24 official languages of the European Union, with content sourced from regulatory summaries on the EU legal database EUR-Lex.
50%