Shimin

/

LLaMA-embeeding

like 0

Evaluation results

• accuracy on MTEB AmazonCounterfactualClassification (en)
  test set self-reported
  84.821
• ap on MTEB AmazonCounterfactualClassification (en)
  test set self-reported
  52.244
• f1 on MTEB AmazonCounterfactualClassification (en)
  test set self-reported
  79.356
• accuracy on MTEB AmazonPolarityClassification
  test set self-reported
  76.880
• ap on MTEB AmazonPolarityClassification
  test set self-reported
  71.756
• f1 on MTEB AmazonPolarityClassification
  test set self-reported
  76.759
• accuracy on MTEB AmazonReviewsClassification (en)
  test set self-reported
  36.716
• f1 on MTEB AmazonReviewsClassification (en)
  test set self-reported
  36.332
• v_measure on MTEB ArxivClusteringS2S
  test set self-reported
  30.154
• map on MTEB AskUbuntuDupQuestions
  test set self-reported
  48.656
• mrr on MTEB AskUbuntuDupQuestions
  test set self-reported
  62.226
• cos_sim_pearson on MTEB BIOSSES
  test set self-reported
  69.187
• cos_sim_spearman on MTEB BIOSSES
  test set self-reported
  67.850
• euclidean_pearson on MTEB BIOSSES
  test set self-reported
  63.366
• euclidean_spearman on MTEB BIOSSES
  test set self-reported
  63.015
• manhattan_pearson on MTEB BIOSSES
  test set self-reported
  63.692
• manhattan_spearman on MTEB BIOSSES
  test set self-reported
  63.597
• v_measure on MTEB BiorxivClusteringS2S
  test set self-reported
  24.294
• accuracy on MTEB EmotionClassification
  test set self-reported
  41.930
• f1 on MTEB EmotionClassification
  test set self-reported
  38.451