Update model card with the current MTEB-ko-retrieval benchmark results

#1
by yjoonjang - opened
NLP & AI - Korea University org
  • Refresh the model card to match the KURE-Reranker-base card: key characteristics, evaluation on the nine MTEB-ko-retrieval subsets (mean nDCG@10 and mean PPS), and per-dataset NDCG@10 / PPS tables.
  • Add assets/pps_vs_ndcg9.png (mean nDCG@10 vs. mean PPS) referenced by the new card.
  • Model weights are unchanged.
hongst changed pull request status to merged

Sign up or log in to comment