mteb/SciDocsRR

SciDocsRR An MTEB dataset Massive Text Embedding Benchmark Ranking of related scientific papers based on their title. Task category t2t Domains Academic, Non-fiction, Written Reference https://allenai.org/data/scidocs How to evaluate on this task You can evaluate an embedding model on this dataset using the following code: import mteb task = mteb.get_tasks(["SciDocsRR"]) evaluator = mteb.MTEB(task) model = mteb.get_model(YOUR_MODEL)… See the full description on the dataset page: https://huggingface.co/datasets/mteb/SciDocsRR.