mteb/

GerDaLIR

GerDaLIR An MTEB dataset Massive Text Embedding Benchmark

GerDaLIR is a legal information retrieval dataset created from the Open Legal Data platform.

Task category t2t

Domains Legal

Reference https://github.com/lavis-nlp/GerDaLIR

Source datasets:

jinaai/ger_da_lir

You can evaluate an embedding model on this dataset using the following code: import mteb

task = mteb.get_task("GerDaLIR") evaluator = mteb.MTEB([task])… See the full description on the dataset page: https://huggingface.co/datasets/mteb/GerDaLIR.