695 B
695 B
German Passage Retrieval (MIRACL)
German dev dataset from the MIRACL challenge.
The dataset consists of 305 queries in the German language.
The expected.tsv
file should contain tab-separated identifiers of passages relevant to the queries (max 10 per query). The identifiers correspond to passages that are located in the German split of the dataset found here.
Directory structure
README.md
— this fileconfig.txt
— configuration filetest-A
— directory with test datatest-A/in.tsv
— input data for the test set