intfloat commited on
Commit
094fd72
1 Parent(s): d6b60af

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +990 -0
README.md CHANGED
@@ -4,6 +4,48 @@ tags:
4
  model-index:
5
  - name: e5-mistral-7b-instruct
6
  results:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  - task:
8
  type: Classification
9
  dataset:
@@ -282,6 +324,27 @@ model-index:
282
  value: 86.67533327742343
283
  - type: manhattan_spearman
284
  value: 85.76099026691983
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
285
  - task:
286
  type: BitextMining
287
  dataset:
@@ -385,6 +448,54 @@ model-index:
385
  metrics:
386
  - type: v_measure
387
  value: 40.23505728593893
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
388
  - task:
389
  type: Retrieval
390
  dataset:
@@ -523,6 +634,199 @@ model-index:
523
  value: 29.008
524
  - type: recall_at_5
525
  value: 35.58
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
526
  - task:
527
  type: Retrieval
528
  dataset:
@@ -592,6 +896,144 @@ model-index:
592
  value: 17.495
593
  - type: recall_at_5
594
  value: 22.235
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
595
  - task:
596
  type: Classification
597
  dataset:
@@ -812,6 +1254,19 @@ model-index:
812
  value: 69.669
813
  - type: recall_at_5
814
  value: 75.604
 
 
 
 
 
 
 
 
 
 
 
 
 
815
  - task:
816
  type: Classification
817
  dataset:
@@ -827,6 +1282,124 @@ model-index:
827
  value: 92.52931921594387
828
  - type: f1
829
  value: 94.77902110732532
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
830
  - task:
831
  type: Retrieval
832
  dataset:
@@ -2378,6 +2951,75 @@ model-index:
2378
  value: 76.86617350369872
2379
  - type: f1
2380
  value: 77.42645654909516
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2381
  - task:
2382
  type: Clustering
2383
  dataset:
@@ -2413,6 +3055,19 @@ model-index:
2413
  value: 32.59661273103955
2414
  - type: mrr
2415
  value: 33.82024242497473
 
 
 
 
 
 
 
 
 
 
 
 
 
2416
  - task:
2417
  type: Retrieval
2418
  dataset:
@@ -2551,6 +3206,118 @@ model-index:
2551
  value: 64.091
2552
  - type: recall_at_5
2553
  value: 74.063
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2554
  - task:
2555
  type: Retrieval
2556
  dataset:
@@ -3446,6 +4213,27 @@ model-index:
3446
  value: 73.94847166379994
3447
  - type: manhattan_spearman
3448
  value: 84.51542547285167
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3449
  - task:
3450
  type: STS
3451
  dataset:
@@ -3656,6 +4444,101 @@ model-index:
3656
  value: 31.469083969291894
3657
  - type: dot_spearman
3658
  value: 31.40325730367437
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3659
  - task:
3660
  type: Retrieval
3661
  dataset:
@@ -5629,6 +6512,28 @@ model-index:
5629
  value: 88.53333333333333
5630
  - type: recall
5631
  value: 91.9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5632
  - task:
5633
  type: Retrieval
5634
  dataset:
@@ -5847,6 +6752,90 @@ model-index:
5847
  value: 87.78807479093082
5848
  - type: max_f1
5849
  value: 80.3221673073403
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5850
  language:
5851
  - en
5852
  license: mit
@@ -5965,3 +6954,4 @@ If you find our paper or models helpful, please consider cite as follows:
5965
  Using this model for inputs longer than 4096 tokens is not recommended.
5966
 
5967
  This model's multilingual capability is still inferior to [multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large) for some cases.
 
 
4
  model-index:
5
  - name: e5-mistral-7b-instruct
6
  results:
7
+ - task:
8
+ type: STS
9
+ dataset:
10
+ type: C-MTEB/AFQMC
11
+ name: MTEB AFQMC
12
+ config: default
13
+ split: validation
14
+ revision: None
15
+ metrics:
16
+ - type: cos_sim_pearson
17
+ value: 37.863226091673866
18
+ - type: cos_sim_spearman
19
+ value: 38.98733013335281
20
+ - type: euclidean_pearson
21
+ value: 37.51783380497874
22
+ - type: euclidean_spearman
23
+ value: 38.98733012753365
24
+ - type: manhattan_pearson
25
+ value: 37.26706888081721
26
+ - type: manhattan_spearman
27
+ value: 38.709750161903834
28
+ - task:
29
+ type: STS
30
+ dataset:
31
+ type: C-MTEB/ATEC
32
+ name: MTEB ATEC
33
+ config: default
34
+ split: test
35
+ revision: None
36
+ metrics:
37
+ - type: cos_sim_pearson
38
+ value: 43.33924583134623
39
+ - type: cos_sim_spearman
40
+ value: 42.84316155158754
41
+ - type: euclidean_pearson
42
+ value: 45.62709879515238
43
+ - type: euclidean_spearman
44
+ value: 42.843155921732404
45
+ - type: manhattan_pearson
46
+ value: 45.4786950991229
47
+ - type: manhattan_spearman
48
+ value: 42.657334751855984
49
  - task:
50
  type: Classification
51
  dataset:
 
324
  value: 86.67533327742343
325
  - type: manhattan_spearman
326
  value: 85.76099026691983
327
+ - task:
328
+ type: STS
329
+ dataset:
330
+ type: C-MTEB/BQ
331
+ name: MTEB BQ
332
+ config: default
333
+ split: test
334
+ revision: None
335
+ metrics:
336
+ - type: cos_sim_pearson
337
+ value: 50.31998888922809
338
+ - type: cos_sim_spearman
339
+ value: 50.6369940530675
340
+ - type: euclidean_pearson
341
+ value: 50.055544636296055
342
+ - type: euclidean_spearman
343
+ value: 50.63699405154838
344
+ - type: manhattan_pearson
345
+ value: 50.00739378036807
346
+ - type: manhattan_spearman
347
+ value: 50.607237418676945
348
  - task:
349
  type: BitextMining
350
  dataset:
 
448
  metrics:
449
  - type: v_measure
450
  value: 40.23505728593893
451
+ - task:
452
+ type: Clustering
453
+ dataset:
454
+ type: C-MTEB/CLSClusteringP2P
455
+ name: MTEB CLSClusteringP2P
456
+ config: default
457
+ split: test
458
+ revision: None
459
+ metrics:
460
+ - type: v_measure
461
+ value: 44.419028279451275
462
+ - task:
463
+ type: Clustering
464
+ dataset:
465
+ type: C-MTEB/CLSClusteringS2S
466
+ name: MTEB CLSClusteringS2S
467
+ config: default
468
+ split: test
469
+ revision: None
470
+ metrics:
471
+ - type: v_measure
472
+ value: 42.5820277929776
473
+ - task:
474
+ type: Reranking
475
+ dataset:
476
+ type: C-MTEB/CMedQAv1-reranking
477
+ name: MTEB CMedQAv1
478
+ config: default
479
+ split: test
480
+ revision: None
481
+ metrics:
482
+ - type: map
483
+ value: 77.67811726152972
484
+ - type: mrr
485
+ value: 80.99003968253969
486
+ - task:
487
+ type: Reranking
488
+ dataset:
489
+ type: C-MTEB/CMedQAv2-reranking
490
+ name: MTEB CMedQAv2
491
+ config: default
492
+ split: test
493
+ revision: None
494
+ metrics:
495
+ - type: map
496
+ value: 78.66055354534922
497
+ - type: mrr
498
+ value: 81.66119047619047
499
  - task:
500
  type: Retrieval
501
  dataset:
 
634
  value: 29.008
635
  - type: recall_at_5
636
  value: 35.58
637
+ - task:
638
+ type: Retrieval
639
+ dataset:
640
+ type: C-MTEB/CmedqaRetrieval
641
+ name: MTEB CmedqaRetrieval
642
+ config: default
643
+ split: dev
644
+ revision: None
645
+ metrics:
646
+ - type: map_at_1
647
+ value: 19.482
648
+ - type: map_at_10
649
+ value: 28.622999999999998
650
+ - type: map_at_100
651
+ value: 30.262
652
+ - type: map_at_1000
653
+ value: 30.432
654
+ - type: map_at_3
655
+ value: 25.647
656
+ - type: map_at_5
657
+ value: 27.128000000000004
658
+ - type: mrr_at_1
659
+ value: 30.408
660
+ - type: mrr_at_10
661
+ value: 37.188
662
+ - type: mrr_at_100
663
+ value: 38.196000000000005
664
+ - type: mrr_at_1000
665
+ value: 38.273
666
+ - type: mrr_at_3
667
+ value: 35.067
668
+ - type: mrr_at_5
669
+ value: 36.124
670
+ - type: ndcg_at_1
671
+ value: 30.408
672
+ - type: ndcg_at_10
673
+ value: 34.215
674
+ - type: ndcg_at_100
675
+ value: 41.349999999999994
676
+ - type: ndcg_at_1000
677
+ value: 44.689
678
+ - type: ndcg_at_3
679
+ value: 30.264999999999997
680
+ - type: ndcg_at_5
681
+ value: 31.572
682
+ - type: precision_at_1
683
+ value: 30.408
684
+ - type: precision_at_10
685
+ value: 7.6770000000000005
686
+ - type: precision_at_100
687
+ value: 1.352
688
+ - type: precision_at_1000
689
+ value: 0.178
690
+ - type: precision_at_3
691
+ value: 17.213
692
+ - type: precision_at_5
693
+ value: 12.198
694
+ - type: recall_at_1
695
+ value: 19.482
696
+ - type: recall_at_10
697
+ value: 42.368
698
+ - type: recall_at_100
699
+ value: 72.694
700
+ - type: recall_at_1000
701
+ value: 95.602
702
+ - type: recall_at_3
703
+ value: 30.101
704
+ - type: recall_at_5
705
+ value: 34.708
706
+ - task:
707
+ type: PairClassification
708
+ dataset:
709
+ type: C-MTEB/CMNLI
710
+ name: MTEB Cmnli
711
+ config: default
712
+ split: validation
713
+ revision: None
714
+ metrics:
715
+ - type: cos_sim_accuracy
716
+ value: 71.16055321707758
717
+ - type: cos_sim_ap
718
+ value: 80.21073839711723
719
+ - type: cos_sim_f1
720
+ value: 72.9740932642487
721
+ - type: cos_sim_precision
722
+ value: 65.53136050623488
723
+ - type: cos_sim_recall
724
+ value: 82.3240589198036
725
+ - type: dot_accuracy
726
+ value: 71.16055321707758
727
+ - type: dot_ap
728
+ value: 80.212299264122
729
+ - type: dot_f1
730
+ value: 72.9740932642487
731
+ - type: dot_precision
732
+ value: 65.53136050623488
733
+ - type: dot_recall
734
+ value: 82.3240589198036
735
+ - type: euclidean_accuracy
736
+ value: 71.16055321707758
737
+ - type: euclidean_ap
738
+ value: 80.21076298680417
739
+ - type: euclidean_f1
740
+ value: 72.9740932642487
741
+ - type: euclidean_precision
742
+ value: 65.53136050623488
743
+ - type: euclidean_recall
744
+ value: 82.3240589198036
745
+ - type: manhattan_accuracy
746
+ value: 70.71557426337944
747
+ - type: manhattan_ap
748
+ value: 79.93448977199749
749
+ - type: manhattan_f1
750
+ value: 72.83962726826877
751
+ - type: manhattan_precision
752
+ value: 62.7407908077053
753
+ - type: manhattan_recall
754
+ value: 86.81318681318682
755
+ - type: max_accuracy
756
+ value: 71.16055321707758
757
+ - type: max_ap
758
+ value: 80.212299264122
759
+ - type: max_f1
760
+ value: 72.9740932642487
761
+ - task:
762
+ type: Retrieval
763
+ dataset:
764
+ type: C-MTEB/CovidRetrieval
765
+ name: MTEB CovidRetrieval
766
+ config: default
767
+ split: dev
768
+ revision: None
769
+ metrics:
770
+ - type: map_at_1
771
+ value: 60.643
772
+ - type: map_at_10
773
+ value: 69.011
774
+ - type: map_at_100
775
+ value: 69.533
776
+ - type: map_at_1000
777
+ value: 69.545
778
+ - type: map_at_3
779
+ value: 67.167
780
+ - type: map_at_5
781
+ value: 68.12700000000001
782
+ - type: mrr_at_1
783
+ value: 60.801
784
+ - type: mrr_at_10
785
+ value: 69.111
786
+ - type: mrr_at_100
787
+ value: 69.6
788
+ - type: mrr_at_1000
789
+ value: 69.611
790
+ - type: mrr_at_3
791
+ value: 67.229
792
+ - type: mrr_at_5
793
+ value: 68.214
794
+ - type: ndcg_at_1
795
+ value: 60.801
796
+ - type: ndcg_at_10
797
+ value: 73.128
798
+ - type: ndcg_at_100
799
+ value: 75.614
800
+ - type: ndcg_at_1000
801
+ value: 75.92
802
+ - type: ndcg_at_3
803
+ value: 69.261
804
+ - type: ndcg_at_5
805
+ value: 70.973
806
+ - type: precision_at_1
807
+ value: 60.801
808
+ - type: precision_at_10
809
+ value: 8.662
810
+ - type: precision_at_100
811
+ value: 0.9860000000000001
812
+ - type: precision_at_1000
813
+ value: 0.101
814
+ - type: precision_at_3
815
+ value: 25.149
816
+ - type: precision_at_5
817
+ value: 15.953999999999999
818
+ - type: recall_at_1
819
+ value: 60.643
820
+ - type: recall_at_10
821
+ value: 85.959
822
+ - type: recall_at_100
823
+ value: 97.576
824
+ - type: recall_at_1000
825
+ value: 100.0
826
+ - type: recall_at_3
827
+ value: 75.184
828
+ - type: recall_at_5
829
+ value: 79.32000000000001
830
  - task:
831
  type: Retrieval
832
  dataset:
 
896
  value: 17.495
897
  - type: recall_at_5
898
  value: 22.235
899
+ - task:
900
+ type: Retrieval
901
+ dataset:
902
+ type: C-MTEB/DuRetrieval
903
+ name: MTEB DuRetrieval
904
+ config: default
905
+ split: dev
906
+ revision: None
907
+ metrics:
908
+ - type: map_at_1
909
+ value: 26.613999999999997
910
+ - type: map_at_10
911
+ value: 79.77300000000001
912
+ - type: map_at_100
913
+ value: 82.71
914
+ - type: map_at_1000
915
+ value: 82.75
916
+ - type: map_at_3
917
+ value: 55.92700000000001
918
+ - type: map_at_5
919
+ value: 70.085
920
+ - type: mrr_at_1
921
+ value: 90.7
922
+ - type: mrr_at_10
923
+ value: 93.438
924
+ - type: mrr_at_100
925
+ value: 93.504
926
+ - type: mrr_at_1000
927
+ value: 93.50699999999999
928
+ - type: mrr_at_3
929
+ value: 93.125
930
+ - type: mrr_at_5
931
+ value: 93.34
932
+ - type: ndcg_at_1
933
+ value: 90.7
934
+ - type: ndcg_at_10
935
+ value: 87.023
936
+ - type: ndcg_at_100
937
+ value: 90.068
938
+ - type: ndcg_at_1000
939
+ value: 90.43299999999999
940
+ - type: ndcg_at_3
941
+ value: 86.339
942
+ - type: ndcg_at_5
943
+ value: 85.013
944
+ - type: precision_at_1
945
+ value: 90.7
946
+ - type: precision_at_10
947
+ value: 41.339999999999996
948
+ - type: precision_at_100
949
+ value: 4.806
950
+ - type: precision_at_1000
951
+ value: 0.48900000000000005
952
+ - type: precision_at_3
953
+ value: 76.983
954
+ - type: precision_at_5
955
+ value: 64.69
956
+ - type: recall_at_1
957
+ value: 26.613999999999997
958
+ - type: recall_at_10
959
+ value: 87.681
960
+ - type: recall_at_100
961
+ value: 97.44699999999999
962
+ - type: recall_at_1000
963
+ value: 99.348
964
+ - type: recall_at_3
965
+ value: 57.809999999999995
966
+ - type: recall_at_5
967
+ value: 74.258
968
+ - task:
969
+ type: Retrieval
970
+ dataset:
971
+ type: C-MTEB/EcomRetrieval
972
+ name: MTEB EcomRetrieval
973
+ config: default
974
+ split: dev
975
+ revision: None
976
+ metrics:
977
+ - type: map_at_1
978
+ value: 30.9
979
+ - type: map_at_10
980
+ value: 40.467
981
+ - type: map_at_100
982
+ value: 41.423
983
+ - type: map_at_1000
984
+ value: 41.463
985
+ - type: map_at_3
986
+ value: 37.25
987
+ - type: map_at_5
988
+ value: 39.31
989
+ - type: mrr_at_1
990
+ value: 30.9
991
+ - type: mrr_at_10
992
+ value: 40.467
993
+ - type: mrr_at_100
994
+ value: 41.423
995
+ - type: mrr_at_1000
996
+ value: 41.463
997
+ - type: mrr_at_3
998
+ value: 37.25
999
+ - type: mrr_at_5
1000
+ value: 39.31
1001
+ - type: ndcg_at_1
1002
+ value: 30.9
1003
+ - type: ndcg_at_10
1004
+ value: 45.957
1005
+ - type: ndcg_at_100
1006
+ value: 50.735
1007
+ - type: ndcg_at_1000
1008
+ value: 51.861999999999995
1009
+ - type: ndcg_at_3
1010
+ value: 39.437
1011
+ - type: ndcg_at_5
1012
+ value: 43.146
1013
+ - type: precision_at_1
1014
+ value: 30.9
1015
+ - type: precision_at_10
1016
+ value: 6.35
1017
+ - type: precision_at_100
1018
+ value: 0.861
1019
+ - type: precision_at_1000
1020
+ value: 0.095
1021
+ - type: precision_at_3
1022
+ value: 15.267
1023
+ - type: precision_at_5
1024
+ value: 10.96
1025
+ - type: recall_at_1
1026
+ value: 30.9
1027
+ - type: recall_at_10
1028
+ value: 63.5
1029
+ - type: recall_at_100
1030
+ value: 86.1
1031
+ - type: recall_at_1000
1032
+ value: 95.1
1033
+ - type: recall_at_3
1034
+ value: 45.800000000000004
1035
+ - type: recall_at_5
1036
+ value: 54.800000000000004
1037
  - task:
1038
  type: Classification
1039
  dataset:
 
1254
  value: 69.669
1255
  - type: recall_at_5
1256
  value: 75.604
1257
+ - task:
1258
+ type: Classification
1259
+ dataset:
1260
+ type: C-MTEB/IFlyTek-classification
1261
+ name: MTEB IFlyTek
1262
+ config: default
1263
+ split: validation
1264
+ revision: None
1265
+ metrics:
1266
+ - type: accuracy
1267
+ value: 45.04809542131589
1268
+ - type: f1
1269
+ value: 37.01181779071118
1270
  - task:
1271
  type: Classification
1272
  dataset:
 
1282
  value: 92.52931921594387
1283
  - type: f1
1284
  value: 94.77902110732532
1285
+ - task:
1286
+ type: Classification
1287
+ dataset:
1288
+ type: C-MTEB/JDReview-classification
1289
+ name: MTEB JDReview
1290
+ config: default
1291
+ split: test
1292
+ revision: None
1293
+ metrics:
1294
+ - type: accuracy
1295
+ value: 85.81613508442777
1296
+ - type: ap
1297
+ value: 52.430320593468394
1298
+ - type: f1
1299
+ value: 79.95467268178068
1300
+ - task:
1301
+ type: STS
1302
+ dataset:
1303
+ type: C-MTEB/LCQMC
1304
+ name: MTEB LCQMC
1305
+ config: default
1306
+ split: test
1307
+ revision: None
1308
+ metrics:
1309
+ - type: cos_sim_pearson
1310
+ value: 71.05801751913393
1311
+ - type: cos_sim_spearman
1312
+ value: 75.47954644971965
1313
+ - type: euclidean_pearson
1314
+ value: 74.27472296759713
1315
+ - type: euclidean_spearman
1316
+ value: 75.47954201369866
1317
+ - type: manhattan_pearson
1318
+ value: 74.30508190186474
1319
+ - type: manhattan_spearman
1320
+ value: 75.51326518159436
1321
+ - task:
1322
+ type: Reranking
1323
+ dataset:
1324
+ type: C-MTEB/Mmarco-reranking
1325
+ name: MTEB MMarcoReranking
1326
+ config: default
1327
+ split: dev
1328
+ revision: None
1329
+ metrics:
1330
+ - type: map
1331
+ value: 24.21110921666315
1332
+ - type: mrr
1333
+ value: 22.863492063492064
1334
+ - task:
1335
+ type: Retrieval
1336
+ dataset:
1337
+ type: C-MTEB/MMarcoRetrieval
1338
+ name: MTEB MMarcoRetrieval
1339
+ config: default
1340
+ split: dev
1341
+ revision: None
1342
+ metrics:
1343
+ - type: map_at_1
1344
+ value: 61.38400000000001
1345
+ - type: map_at_10
1346
+ value: 70.895
1347
+ - type: map_at_100
1348
+ value: 71.314
1349
+ - type: map_at_1000
1350
+ value: 71.331
1351
+ - type: map_at_3
1352
+ value: 69.016
1353
+ - type: map_at_5
1354
+ value: 70.179
1355
+ - type: mrr_at_1
1356
+ value: 63.481
1357
+ - type: mrr_at_10
1358
+ value: 71.543
1359
+ - type: mrr_at_100
1360
+ value: 71.91300000000001
1361
+ - type: mrr_at_1000
1362
+ value: 71.928
1363
+ - type: mrr_at_3
1364
+ value: 69.90899999999999
1365
+ - type: mrr_at_5
1366
+ value: 70.907
1367
+ - type: ndcg_at_1
1368
+ value: 63.481
1369
+ - type: ndcg_at_10
1370
+ value: 74.833
1371
+ - type: ndcg_at_100
1372
+ value: 76.705
1373
+ - type: ndcg_at_1000
1374
+ value: 77.13600000000001
1375
+ - type: ndcg_at_3
1376
+ value: 71.236
1377
+ - type: ndcg_at_5
1378
+ value: 73.199
1379
+ - type: precision_at_1
1380
+ value: 63.481
1381
+ - type: precision_at_10
1382
+ value: 9.179
1383
+ - type: precision_at_100
1384
+ value: 1.011
1385
+ - type: precision_at_1000
1386
+ value: 0.105
1387
+ - type: precision_at_3
1388
+ value: 27.044
1389
+ - type: precision_at_5
1390
+ value: 17.272000000000002
1391
+ - type: recall_at_1
1392
+ value: 61.38400000000001
1393
+ - type: recall_at_10
1394
+ value: 86.318
1395
+ - type: recall_at_100
1396
+ value: 94.786
1397
+ - type: recall_at_1000
1398
+ value: 98.14500000000001
1399
+ - type: recall_at_3
1400
+ value: 76.717
1401
+ - type: recall_at_5
1402
+ value: 81.416
1403
  - task:
1404
  type: Retrieval
1405
  dataset:
 
2951
  value: 76.86617350369872
2952
  - type: f1
2953
  value: 77.42645654909516
2954
+ - task:
2955
+ type: Retrieval
2956
+ dataset:
2957
+ type: C-MTEB/MedicalRetrieval
2958
+ name: MTEB MedicalRetrieval
2959
+ config: default
2960
+ split: dev
2961
+ revision: None
2962
+ metrics:
2963
+ - type: map_at_1
2964
+ value: 44.6
2965
+ - type: map_at_10
2966
+ value: 50.019000000000005
2967
+ - type: map_at_100
2968
+ value: 50.611
2969
+ - type: map_at_1000
2970
+ value: 50.67
2971
+ - type: map_at_3
2972
+ value: 48.699999999999996
2973
+ - type: map_at_5
2974
+ value: 49.455
2975
+ - type: mrr_at_1
2976
+ value: 44.800000000000004
2977
+ - type: mrr_at_10
2978
+ value: 50.119
2979
+ - type: mrr_at_100
2980
+ value: 50.711
2981
+ - type: mrr_at_1000
2982
+ value: 50.77
2983
+ - type: mrr_at_3
2984
+ value: 48.8
2985
+ - type: mrr_at_5
2986
+ value: 49.555
2987
+ - type: ndcg_at_1
2988
+ value: 44.6
2989
+ - type: ndcg_at_10
2990
+ value: 52.754
2991
+ - type: ndcg_at_100
2992
+ value: 55.935
2993
+ - type: ndcg_at_1000
2994
+ value: 57.607
2995
+ - type: ndcg_at_3
2996
+ value: 50.012
2997
+ - type: ndcg_at_5
2998
+ value: 51.393
2999
+ - type: precision_at_1
3000
+ value: 44.6
3001
+ - type: precision_at_10
3002
+ value: 6.140000000000001
3003
+ - type: precision_at_100
3004
+ value: 0.77
3005
+ - type: precision_at_1000
3006
+ value: 0.09
3007
+ - type: precision_at_3
3008
+ value: 17.933
3009
+ - type: precision_at_5
3010
+ value: 11.44
3011
+ - type: recall_at_1
3012
+ value: 44.6
3013
+ - type: recall_at_10
3014
+ value: 61.4
3015
+ - type: recall_at_100
3016
+ value: 77.0
3017
+ - type: recall_at_1000
3018
+ value: 90.4
3019
+ - type: recall_at_3
3020
+ value: 53.800000000000004
3021
+ - type: recall_at_5
3022
+ value: 57.199999999999996
3023
  - task:
3024
  type: Clustering
3025
  dataset:
 
3055
  value: 32.59661273103955
3056
  - type: mrr
3057
  value: 33.82024242497473
3058
+ - task:
3059
+ type: Classification
3060
+ dataset:
3061
+ type: C-MTEB/MultilingualSentiment-classification
3062
+ name: MTEB MultilingualSentiment
3063
+ config: default
3064
+ split: validation
3065
+ revision: None
3066
+ metrics:
3067
+ - type: accuracy
3068
+ value: 73.31333333333335
3069
+ - type: f1
3070
+ value: 73.0873466527602
3071
  - task:
3072
  type: Retrieval
3073
  dataset:
 
3206
  value: 64.091
3207
  - type: recall_at_5
3208
  value: 74.063
3209
+ - task:
3210
+ type: PairClassification
3211
+ dataset:
3212
+ type: C-MTEB/OCNLI
3213
+ name: MTEB Ocnli
3214
+ config: default
3215
+ split: validation
3216
+ revision: None
3217
+ metrics:
3218
+ - type: cos_sim_accuracy
3219
+ value: 61.18029236599891
3220
+ - type: cos_sim_ap
3221
+ value: 64.18398769398412
3222
+ - type: cos_sim_f1
3223
+ value: 67.96347757046446
3224
+ - type: cos_sim_precision
3225
+ value: 54.4529262086514
3226
+ - type: cos_sim_recall
3227
+ value: 90.3907074973601
3228
+ - type: dot_accuracy
3229
+ value: 61.18029236599891
3230
+ - type: dot_ap
3231
+ value: 64.18393484706077
3232
+ - type: dot_f1
3233
+ value: 67.96347757046446
3234
+ - type: dot_precision
3235
+ value: 54.4529262086514
3236
+ - type: dot_recall
3237
+ value: 90.3907074973601
3238
+ - type: euclidean_accuracy
3239
+ value: 61.18029236599891
3240
+ - type: euclidean_ap
3241
+ value: 64.18395024821486
3242
+ - type: euclidean_f1
3243
+ value: 67.96347757046446
3244
+ - type: euclidean_precision
3245
+ value: 54.4529262086514
3246
+ - type: euclidean_recall
3247
+ value: 90.3907074973601
3248
+ - type: manhattan_accuracy
3249
+ value: 61.451001624255554
3250
+ - type: manhattan_ap
3251
+ value: 64.38232708763513
3252
+ - type: manhattan_f1
3253
+ value: 68.05860805860804
3254
+ - type: manhattan_precision
3255
+ value: 52.10319685922602
3256
+ - type: manhattan_recall
3257
+ value: 98.09926082365365
3258
+ - type: max_accuracy
3259
+ value: 61.451001624255554
3260
+ - type: max_ap
3261
+ value: 64.38232708763513
3262
+ - type: max_f1
3263
+ value: 68.05860805860804
3264
+ - task:
3265
+ type: Classification
3266
+ dataset:
3267
+ type: C-MTEB/OnlineShopping-classification
3268
+ name: MTEB OnlineShopping
3269
+ config: default
3270
+ split: test
3271
+ revision: None
3272
+ metrics:
3273
+ - type: accuracy
3274
+ value: 92.19000000000001
3275
+ - type: ap
3276
+ value: 89.73918431886767
3277
+ - type: f1
3278
+ value: 92.17175032574507
3279
+ - task:
3280
+ type: STS
3281
+ dataset:
3282
+ type: C-MTEB/PAWSX
3283
+ name: MTEB PAWSX
3284
+ config: default
3285
+ split: test
3286
+ revision: None
3287
+ metrics:
3288
+ - type: cos_sim_pearson
3289
+ value: 15.079320253752224
3290
+ - type: cos_sim_spearman
3291
+ value: 16.813772504404263
3292
+ - type: euclidean_pearson
3293
+ value: 19.476541162041762
3294
+ - type: euclidean_spearman
3295
+ value: 16.813772498098782
3296
+ - type: manhattan_pearson
3297
+ value: 19.497429832915277
3298
+ - type: manhattan_spearman
3299
+ value: 16.869600674180607
3300
+ - task:
3301
+ type: STS
3302
+ dataset:
3303
+ type: C-MTEB/QBQTC
3304
+ name: MTEB QBQTC
3305
+ config: default
3306
+ split: test
3307
+ revision: None
3308
+ metrics:
3309
+ - type: cos_sim_pearson
3310
+ value: 30.36139599797913
3311
+ - type: cos_sim_spearman
3312
+ value: 31.80296402851347
3313
+ - type: euclidean_pearson
3314
+ value: 30.10387888252793
3315
+ - type: euclidean_spearman
3316
+ value: 31.80297780103808
3317
+ - type: manhattan_pearson
3318
+ value: 30.86720382849436
3319
+ - type: manhattan_spearman
3320
+ value: 32.70491131366606
3321
  - task:
3322
  type: Retrieval
3323
  dataset:
 
4213
  value: 73.94847166379994
4214
  - type: manhattan_spearman
4215
  value: 84.51542547285167
4216
+ - task:
4217
+ type: STS
4218
+ dataset:
4219
+ type: C-MTEB/STSB
4220
+ name: MTEB STSB
4221
+ config: default
4222
+ split: test
4223
+ revision: None
4224
+ metrics:
4225
+ - type: cos_sim_pearson
4226
+ value: 81.78690149086131
4227
+ - type: cos_sim_spearman
4228
+ value: 81.81202616916873
4229
+ - type: euclidean_pearson
4230
+ value: 80.98792254251062
4231
+ - type: euclidean_spearman
4232
+ value: 81.81202616916873
4233
+ - type: manhattan_pearson
4234
+ value: 81.46953021346732
4235
+ - type: manhattan_spearman
4236
+ value: 82.34259562492315
4237
  - task:
4238
  type: STS
4239
  dataset:
 
4444
  value: 31.469083969291894
4445
  - type: dot_spearman
4446
  value: 31.40325730367437
4447
+ - task:
4448
+ type: Reranking
4449
+ dataset:
4450
+ type: C-MTEB/T2Reranking
4451
+ name: MTEB T2Reranking
4452
+ config: default
4453
+ split: dev
4454
+ revision: None
4455
+ metrics:
4456
+ - type: map
4457
+ value: 66.90300783402137
4458
+ - type: mrr
4459
+ value: 77.06451972574179
4460
+ - task:
4461
+ type: Retrieval
4462
+ dataset:
4463
+ type: C-MTEB/T2Retrieval
4464
+ name: MTEB T2Retrieval
4465
+ config: default
4466
+ split: dev
4467
+ revision: None
4468
+ metrics:
4469
+ - type: map_at_1
4470
+ value: 25.82
4471
+ - type: map_at_10
4472
+ value: 72.32300000000001
4473
+ - type: map_at_100
4474
+ value: 76.198
4475
+ - type: map_at_1000
4476
+ value: 76.281
4477
+ - type: map_at_3
4478
+ value: 50.719
4479
+ - type: map_at_5
4480
+ value: 62.326
4481
+ - type: mrr_at_1
4482
+ value: 86.599
4483
+ - type: mrr_at_10
4484
+ value: 89.751
4485
+ - type: mrr_at_100
4486
+ value: 89.876
4487
+ - type: mrr_at_1000
4488
+ value: 89.88000000000001
4489
+ - type: mrr_at_3
4490
+ value: 89.151
4491
+ - type: mrr_at_5
4492
+ value: 89.519
4493
+ - type: ndcg_at_1
4494
+ value: 86.599
4495
+ - type: ndcg_at_10
4496
+ value: 80.676
4497
+ - type: ndcg_at_100
4498
+ value: 85.03
4499
+ - type: ndcg_at_1000
4500
+ value: 85.854
4501
+ - type: ndcg_at_3
4502
+ value: 82.057
4503
+ - type: ndcg_at_5
4504
+ value: 80.537
4505
+ - type: precision_at_1
4506
+ value: 86.599
4507
+ - type: precision_at_10
4508
+ value: 40.373
4509
+ - type: precision_at_100
4510
+ value: 4.95
4511
+ - type: precision_at_1000
4512
+ value: 0.514
4513
+ - type: precision_at_3
4514
+ value: 71.918
4515
+ - type: precision_at_5
4516
+ value: 60.246
4517
+ - type: recall_at_1
4518
+ value: 25.82
4519
+ - type: recall_at_10
4520
+ value: 79.905
4521
+ - type: recall_at_100
4522
+ value: 93.88499999999999
4523
+ - type: recall_at_1000
4524
+ value: 98.073
4525
+ - type: recall_at_3
4526
+ value: 52.623
4527
+ - type: recall_at_5
4528
+ value: 66.233
4529
+ - task:
4530
+ type: Classification
4531
+ dataset:
4532
+ type: C-MTEB/TNews-classification
4533
+ name: MTEB TNews
4534
+ config: default
4535
+ split: validation
4536
+ revision: None
4537
+ metrics:
4538
+ - type: accuracy
4539
+ value: 47.050000000000004
4540
+ - type: f1
4541
+ value: 45.704071498353294
4542
  - task:
4543
  type: Retrieval
4544
  dataset:
 
6512
  value: 88.53333333333333
6513
  - type: recall
6514
  value: 91.9
6515
+ - task:
6516
+ type: Clustering
6517
+ dataset:
6518
+ type: C-MTEB/ThuNewsClusteringP2P
6519
+ name: MTEB ThuNewsClusteringP2P
6520
+ config: default
6521
+ split: test
6522
+ revision: None
6523
+ metrics:
6524
+ - type: v_measure
6525
+ value: 64.68478289806511
6526
+ - task:
6527
+ type: Clustering
6528
+ dataset:
6529
+ type: C-MTEB/ThuNewsClusteringS2S
6530
+ name: MTEB ThuNewsClusteringS2S
6531
+ config: default
6532
+ split: test
6533
+ revision: None
6534
+ metrics:
6535
+ - type: v_measure
6536
+ value: 57.53010296184097
6537
  - task:
6538
  type: Retrieval
6539
  dataset:
 
6752
  value: 87.78807479093082
6753
  - type: max_f1
6754
  value: 80.3221673073403
6755
+ - task:
6756
+ type: Retrieval
6757
+ dataset:
6758
+ type: C-MTEB/VideoRetrieval
6759
+ name: MTEB VideoRetrieval
6760
+ config: default
6761
+ split: dev
6762
+ revision: None
6763
+ metrics:
6764
+ - type: map_at_1
6765
+ value: 32.4
6766
+ - type: map_at_10
6767
+ value: 40.961999999999996
6768
+ - type: map_at_100
6769
+ value: 41.660000000000004
6770
+ - type: map_at_1000
6771
+ value: 41.721000000000004
6772
+ - type: map_at_3
6773
+ value: 38.550000000000004
6774
+ - type: map_at_5
6775
+ value: 40.06
6776
+ - type: mrr_at_1
6777
+ value: 32.4
6778
+ - type: mrr_at_10
6779
+ value: 40.961999999999996
6780
+ - type: mrr_at_100
6781
+ value: 41.660000000000004
6782
+ - type: mrr_at_1000
6783
+ value: 41.721000000000004
6784
+ - type: mrr_at_3
6785
+ value: 38.550000000000004
6786
+ - type: mrr_at_5
6787
+ value: 40.06
6788
+ - type: ndcg_at_1
6789
+ value: 32.4
6790
+ - type: ndcg_at_10
6791
+ value: 45.388
6792
+ - type: ndcg_at_100
6793
+ value: 49.012
6794
+ - type: ndcg_at_1000
6795
+ value: 50.659
6796
+ - type: ndcg_at_3
6797
+ value: 40.47
6798
+ - type: ndcg_at_5
6799
+ value: 43.232
6800
+ - type: precision_at_1
6801
+ value: 32.4
6802
+ - type: precision_at_10
6803
+ value: 5.94
6804
+ - type: precision_at_100
6805
+ value: 0.769
6806
+ - type: precision_at_1000
6807
+ value: 0.09
6808
+ - type: precision_at_3
6809
+ value: 15.333
6810
+ - type: precision_at_5
6811
+ value: 10.56
6812
+ - type: recall_at_1
6813
+ value: 32.4
6814
+ - type: recall_at_10
6815
+ value: 59.4
6816
+ - type: recall_at_100
6817
+ value: 76.9
6818
+ - type: recall_at_1000
6819
+ value: 90.0
6820
+ - type: recall_at_3
6821
+ value: 46.0
6822
+ - type: recall_at_5
6823
+ value: 52.800000000000004
6824
+ - task:
6825
+ type: Classification
6826
+ dataset:
6827
+ type: C-MTEB/waimai-classification
6828
+ name: MTEB Waimai
6829
+ config: default
6830
+ split: test
6831
+ revision: None
6832
+ metrics:
6833
+ - type: accuracy
6834
+ value: 86.94000000000001
6835
+ - type: ap
6836
+ value: 70.57373468481975
6837
+ - type: f1
6838
+ value: 85.26264784928323
6839
  language:
6840
  - en
6841
  license: mit
 
6954
  Using this model for inputs longer than 4096 tokens is not recommended.
6955
 
6956
  This model's multilingual capability is still inferior to [multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large) for some cases.
6957
+