metadata
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:502912
- loss:MarginMSELoss
- mteb
model-index:
- name: XLM-RoBERTa-base-MSMARCO
results:
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (ar)
config: ar
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 17.861
- type: map_at_10
value: 28.733999999999998
- type: map_at_100
value: 30.482
- type: map_at_1000
value: 30.616
- type: map_at_20
value: 29.717
- type: map_at_3
value: 24.617
- type: map_at_5
value: 26.654
- type: mrr_at_1
value: 26.900000000000002
- type: mrr_at_10
value: 37.138
- type: mrr_at_100
value: 38.157999999999994
- type: mrr_at_1000
value: 38.204
- type: mrr_at_20
value: 37.808
- type: mrr_at_3
value: 33.717000000000006
- type: mrr_at_5
value: 35.632000000000005
- type: ndcg_at_1
value: 26.900000000000002
- type: ndcg_at_10
value: 36.125
- type: ndcg_at_100
value: 43.043
- type: ndcg_at_1000
value: 45.611000000000004
- type: ndcg_at_20
value: 39.055
- type: ndcg_at_3
value: 29.372
- type: ndcg_at_5
value: 32.015
- type: precision_at_1
value: 26.900000000000002
- type: precision_at_10
value: 8.59
- type: precision_at_100
value: 1.444
- type: precision_at_1000
value: 0.181
- type: precision_at_20
value: 5.295
- type: precision_at_3
value: 16.933
- type: precision_at_5
value: 12.839999999999998
- type: recall_at_1
value: 17.861
- type: recall_at_10
value: 49.175999999999995
- type: recall_at_100
value: 76.28399999999999
- type: recall_at_1000
value: 92.905
- type: recall_at_20
value: 58.755
- type: recall_at_3
value: 31.334
- type: recall_at_5
value: 38.217
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (arabic)
config: arabic
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 16.744
- type: map_at_10
value: 25.447999999999997
- type: map_at_100
value: 26.296999999999997
- type: map_at_1000
value: 26.369
- type: map_at_20
value: 25.951
- type: map_at_3
value: 22.826
- type: map_at_5
value: 24.309
- type: mrr_at_1
value: 17.669
- type: mrr_at_10
value: 26.711000000000002
- type: mrr_at_100
value: 27.534
- type: mrr_at_1000
value: 27.589000000000002
- type: mrr_at_20
value: 27.211999999999996
- type: mrr_at_3
value: 24.052
- type: mrr_at_5
value: 25.532
- type: ndcg_at_1
value: 17.669
- type: ndcg_at_10
value: 30.53
- type: ndcg_at_100
value: 34.823
- type: ndcg_at_1000
value: 36.624
- type: ndcg_at_20
value: 32.323
- type: ndcg_at_3
value: 25.119000000000003
- type: ndcg_at_5
value: 27.791
- type: precision_at_1
value: 17.669
- type: precision_at_10
value: 4.8660000000000005
- type: precision_at_100
value: 0.7230000000000001
- type: precision_at_1000
value: 0.09
- type: precision_at_20
value: 2.826
- type: precision_at_3
value: 10.823
- type: precision_at_5
value: 7.9
- type: recall_at_1
value: 16.744
- type: recall_at_10
value: 45.205
- type: recall_at_100
value: 65.094
- type: recall_at_1000
value: 78.739
- type: recall_at_20
value: 52.066
- type: recall_at_3
value: 30.574
- type: recall_at_5
value: 36.986999999999995
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (bn)
config: bn
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 15.962000000000002
- type: map_at_10
value: 24.746000000000002
- type: map_at_100
value: 26.306
- type: map_at_1000
value: 26.451
- type: map_at_20
value: 25.507
- type: map_at_3
value: 20.810000000000002
- type: map_at_5
value: 23.095
- type: mrr_at_1
value: 26.521
- type: mrr_at_10
value: 36.268
- type: mrr_at_100
value: 36.986999999999995
- type: mrr_at_1000
value: 37.047000000000004
- type: mrr_at_20
value: 36.611
- type: mrr_at_3
value: 32.644
- type: mrr_at_5
value: 35.223
- type: ndcg_at_1
value: 26.521
- type: ndcg_at_10
value: 32.038
- type: ndcg_at_100
value: 38.577
- type: ndcg_at_1000
value: 41.449999999999996
- type: ndcg_at_20
value: 34.086
- type: ndcg_at_3
value: 25.515
- type: ndcg_at_5
value: 28.860999999999997
- type: precision_at_1
value: 26.521
- type: precision_at_10
value: 8.224
- type: precision_at_100
value: 1.426
- type: precision_at_1000
value: 0.183
- type: precision_at_20
value: 4.854
- type: precision_at_3
value: 15.328
- type: precision_at_5
value: 12.701
- type: recall_at_1
value: 15.962000000000002
- type: recall_at_10
value: 41.573
- type: recall_at_100
value: 67.963
- type: recall_at_1000
value: 87.077
- type: recall_at_20
value: 48.036
- type: recall_at_3
value: 25.457
- type: recall_at_5
value: 33.143
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (bengali)
config: bengali
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 14.865
- type: map_at_10
value: 24.122
- type: map_at_100
value: 25.345000000000002
- type: map_at_1000
value: 25.413000000000004
- type: map_at_20
value: 24.887999999999998
- type: map_at_3
value: 19.97
- type: map_at_5
value: 22.898
- type: mrr_at_1
value: 17.116999999999997
- type: mrr_at_10
value: 26.115
- type: mrr_at_100
value: 27.194000000000003
- type: mrr_at_1000
value: 27.243000000000002
- type: mrr_at_20
value: 26.828999999999997
- type: mrr_at_3
value: 21.922
- type: mrr_at_5
value: 24.985
- type: ndcg_at_1
value: 17.116999999999997
- type: ndcg_at_10
value: 29.918
- type: ndcg_at_100
value: 35.658
- type: ndcg_at_1000
value: 37.349
- type: ndcg_at_20
value: 32.74
- type: ndcg_at_3
value: 21.807000000000002
- type: ndcg_at_5
value: 27.128999999999998
- type: precision_at_1
value: 17.116999999999997
- type: precision_at_10
value: 5.405
- type: precision_at_100
value: 0.8380000000000001
- type: precision_at_1000
value: 0.099
- type: precision_at_20
value: 3.288
- type: precision_at_3
value: 9.91
- type: precision_at_5
value: 8.828999999999999
- type: recall_at_1
value: 14.865
- type: recall_at_10
value: 45.946
- type: recall_at_100
value: 72.072
- type: recall_at_1000
value: 85.135
- type: recall_at_20
value: 57.206999999999994
- type: recall_at_3
value: 25.224999999999998
- type: recall_at_5
value: 37.838
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (de)
config: de
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 10.285
- type: map_at_10
value: 22.831000000000003
- type: map_at_100
value: 25.868000000000002
- type: map_at_1000
value: 26.029999999999998
- type: map_at_20
value: 24.3
- type: map_at_3
value: 16.539
- type: map_at_5
value: 19.593
- type: mrr_at_1
value: 23.607
- type: mrr_at_10
value: 36.899
- type: mrr_at_100
value: 37.958
- type: mrr_at_1000
value: 37.987
- type: mrr_at_20
value: 37.454
- type: mrr_at_3
value: 32.732
- type: mrr_at_5
value: 35.126000000000005
- type: ndcg_at_1
value: 23.607
- type: ndcg_at_10
value: 32.774
- type: ndcg_at_100
value: 43.553999999999995
- type: ndcg_at_1000
value: 45.604
- type: ndcg_at_20
value: 36.408
- type: ndcg_at_3
value: 23.846
- type: ndcg_at_5
value: 26.954
- type: precision_at_1
value: 23.607
- type: precision_at_10
value: 11.639
- type: precision_at_100
value: 2.2689999999999997
- type: precision_at_1000
value: 0.258
- type: precision_at_20
value: 7.327999999999999
- type: precision_at_3
value: 18.251
- type: precision_at_5
value: 15.344
- type: recall_at_1
value: 10.285
- type: recall_at_10
value: 46.227000000000004
- type: recall_at_100
value: 85.478
- type: recall_at_1000
value: 97.52
- type: recall_at_20
value: 56.738
- type: recall_at_3
value: 22.633
- type: recall_at_5
value: 31.898
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (en)
config: en
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 11.625
- type: map_at_10
value: 22.033
- type: map_at_100
value: 25.343
- type: map_at_1000
value: 25.555
- type: map_at_20
value: 23.77
- type: map_at_3
value: 17.163
- type: map_at_5
value: 19.431
- type: mrr_at_1
value: 24.781
- type: mrr_at_10
value: 35.998000000000005
- type: mrr_at_100
value: 37.345
- type: mrr_at_1000
value: 37.376
- type: mrr_at_20
value: 36.912
- type: mrr_at_3
value: 32.144
- type: mrr_at_5
value: 34.184
- type: ndcg_at_1
value: 24.781
- type: ndcg_at_10
value: 30.608999999999998
- type: ndcg_at_100
value: 42.407000000000004
- type: ndcg_at_1000
value: 45.1
- type: ndcg_at_20
value: 35.188
- type: ndcg_at_3
value: 24.271
- type: ndcg_at_5
value: 26.090000000000003
- type: precision_at_1
value: 24.781
- type: precision_at_10
value: 10.388
- type: precision_at_100
value: 2.35
- type: precision_at_1000
value: 0.28400000000000003
- type: precision_at_20
value: 7.140000000000001
- type: precision_at_3
value: 17.438000000000002
- type: precision_at_5
value: 14.043
- type: recall_at_1
value: 11.625
- type: recall_at_10
value: 40.717
- type: recall_at_100
value: 82.492
- type: recall_at_1000
value: 97.495
- type: recall_at_20
value: 54.208
- type: recall_at_3
value: 21.747
- type: recall_at_5
value: 28.74
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (english)
config: english
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 6.765000000000001
- type: map_at_10
value: 13.469999999999999
- type: map_at_100
value: 14.575
- type: map_at_1000
value: 14.668000000000001
- type: map_at_20
value: 14.055000000000001
- type: map_at_3
value: 10.853
- type: map_at_5
value: 12.251
- type: mrr_at_1
value: 8.602
- type: mrr_at_10
value: 15.595999999999998
- type: mrr_at_100
value: 16.624
- type: mrr_at_1000
value: 16.704
- type: mrr_at_20
value: 16.169
- type: mrr_at_3
value: 12.881
- type: mrr_at_5
value: 14.393
- type: ndcg_at_1
value: 8.602
- type: ndcg_at_10
value: 18.107
- type: ndcg_at_100
value: 23.851
- type: ndcg_at_1000
value: 26.379
- type: ndcg_at_20
value: 20.182
- type: ndcg_at_3
value: 12.681000000000001
- type: ndcg_at_5
value: 15.190000000000001
- type: precision_at_1
value: 8.602
- type: precision_at_10
value: 3.6830000000000003
- type: precision_at_100
value: 0.7040000000000001
- type: precision_at_1000
value: 0.095
- type: precision_at_20
value: 2.332
- type: precision_at_3
value: 6.407
- type: precision_at_5
value: 5.188000000000001
- type: recall_at_1
value: 6.765000000000001
- type: recall_at_10
value: 30.511
- type: recall_at_100
value: 57.303000000000004
- type: recall_at_1000
value: 76.568
- type: recall_at_20
value: 38.284
- type: recall_at_3
value: 16.084
- type: recall_at_5
value: 21.886
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (fa)
config: fa
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 15.268
- type: map_at_10
value: 27.929
- type: map_at_100
value: 30.323
- type: map_at_1000
value: 30.433
- type: map_at_20
value: 29.247
- type: map_at_3
value: 22.684
- type: map_at_5
value: 25.453
- type: mrr_at_1
value: 24.684
- type: mrr_at_10
value: 36.925999999999995
- type: mrr_at_100
value: 38.031
- type: mrr_at_1000
value: 38.063
- type: mrr_at_20
value: 37.62
- type: mrr_at_3
value: 33.07
- type: mrr_at_5
value: 35.467
- type: ndcg_at_1
value: 24.684
- type: ndcg_at_10
value: 36.153999999999996
- type: ndcg_at_100
value: 44.828
- type: ndcg_at_1000
value: 46.482
- type: ndcg_at_20
value: 39.678999999999995
- type: ndcg_at_3
value: 28.223
- type: ndcg_at_5
value: 31.684
- type: precision_at_1
value: 24.684
- type: precision_at_10
value: 9.778
- type: precision_at_100
value: 1.7610000000000001
- type: precision_at_1000
value: 0.199
- type: precision_at_20
value: 6.178999999999999
- type: precision_at_3
value: 17.563000000000002
- type: precision_at_5
value: 14.272000000000002
- type: recall_at_1
value: 15.268
- type: recall_at_10
value: 50.39
- type: recall_at_100
value: 83.39800000000001
- type: recall_at_1000
value: 93.589
- type: recall_at_20
value: 61.543000000000006
- type: recall_at_3
value: 29.754
- type: recall_at_5
value: 38.504
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (fi)
config: fi
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 20.075000000000003
- type: map_at_10
value: 33.783
- type: map_at_100
value: 35.548
- type: map_at_1000
value: 35.650999999999996
- type: map_at_20
value: 34.824
- type: map_at_3
value: 29.069
- type: map_at_5
value: 31.887999999999998
- type: mrr_at_1
value: 32.9
- type: mrr_at_10
value: 44.633
- type: mrr_at_100
value: 45.491
- type: mrr_at_1000
value: 45.518
- type: mrr_at_20
value: 45.184999999999995
- type: mrr_at_3
value: 41.533
- type: mrr_at_5
value: 43.338
- type: ndcg_at_1
value: 32.9
- type: ndcg_at_10
value: 42.068
- type: ndcg_at_100
value: 48.836
- type: ndcg_at_1000
value: 50.849999999999994
- type: ndcg_at_20
value: 45.013999999999996
- type: ndcg_at_3
value: 34.382000000000005
- type: ndcg_at_5
value: 38.169
- type: precision_at_1
value: 32.9
- type: precision_at_10
value: 10.02
- type: precision_at_100
value: 1.545
- type: precision_at_1000
value: 0.182
- type: precision_at_20
value: 5.955
- type: precision_at_3
value: 20.9
- type: precision_at_5
value: 15.98
- type: recall_at_1
value: 20.075000000000003
- type: recall_at_10
value: 55.042
- type: recall_at_100
value: 81.609
- type: recall_at_1000
value: 95.048
- type: recall_at_20
value: 64.791
- type: recall_at_3
value: 36.617
- type: recall_at_5
value: 44.905
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (finnish)
config: finnish
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 13.489999999999998
- type: map_at_10
value: 20.253
- type: map_at_100
value: 21.201
- type: map_at_1000
value: 21.285999999999998
- type: map_at_20
value: 20.796
- type: map_at_3
value: 17.869
- type: map_at_5
value: 19.217000000000002
- type: mrr_at_1
value: 14.514
- type: mrr_at_10
value: 21.369
- type: mrr_at_100
value: 22.273
- type: mrr_at_1000
value: 22.346
- type: mrr_at_20
value: 21.892
- type: mrr_at_3
value: 18.913
- type: mrr_at_5
value: 20.336000000000002
- type: ndcg_at_1
value: 14.514
- type: ndcg_at_10
value: 24.426000000000002
- type: ndcg_at_100
value: 29.253
- type: ndcg_at_1000
value: 31.752000000000002
- type: ndcg_at_20
value: 26.36
- type: ndcg_at_3
value: 19.523
- type: ndcg_at_5
value: 21.986
- type: precision_at_1
value: 14.514
- type: precision_at_10
value: 4.011
- type: precision_at_100
value: 0.6629999999999999
- type: precision_at_1000
value: 0.09
- type: precision_at_20
value: 2.444
- type: precision_at_3
value: 8.373
- type: precision_at_5
value: 6.332
- type: recall_at_1
value: 13.489999999999998
- type: recall_at_10
value: 36.43
- type: recall_at_100
value: 59.024
- type: recall_at_1000
value: 78.668
- type: recall_at_20
value: 43.793
- type: recall_at_3
value: 23.339
- type: recall_at_5
value: 29.213
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (fr)
config: fr
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 7.768
- type: map_at_10
value: 17.591
- type: map_at_100
value: 20.544999999999998
- type: map_at_1000
value: 20.673
- type: map_at_20
value: 19.131
- type: map_at_3
value: 12.567999999999998
- type: map_at_5
value: 14.846
- type: mrr_at_1
value: 13.994000000000002
- type: mrr_at_10
value: 25.740000000000002
- type: mrr_at_100
value: 27.156999999999996
- type: mrr_at_1000
value: 27.189999999999998
- type: mrr_at_20
value: 26.61
- type: mrr_at_3
value: 20.554
- type: mrr_at_5
value: 23.163
- type: ndcg_at_1
value: 13.994000000000002
- type: ndcg_at_10
value: 25.929000000000002
- type: ndcg_at_100
value: 37.244
- type: ndcg_at_1000
value: 39.211
- type: ndcg_at_20
value: 30.314000000000004
- type: ndcg_at_3
value: 16.329
- type: ndcg_at_5
value: 19.826
- type: precision_at_1
value: 13.994000000000002
- type: precision_at_10
value: 8.192
- type: precision_at_100
value: 1.799
- type: precision_at_1000
value: 0.207
- type: precision_at_20
value: 5.583
- type: precision_at_3
value: 11.079
- type: precision_at_5
value: 10.204
- type: recall_at_1
value: 7.768
- type: recall_at_10
value: 41.516999999999996
- type: recall_at_100
value: 85.148
- type: recall_at_1000
value: 97.342
- type: recall_at_20
value: 55.696
- type: recall_at_3
value: 17.559
- type: recall_at_5
value: 25.128
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (hi)
config: hi
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 12.864
- type: map_at_10
value: 23.992
- type: map_at_100
value: 26.007
- type: map_at_1000
value: 26.157999999999998
- type: map_at_20
value: 25.130000000000003
- type: map_at_3
value: 19.867
- type: map_at_5
value: 21.657
- type: mrr_at_1
value: 25.714
- type: mrr_at_10
value: 36.374
- type: mrr_at_100
value: 37.399
- type: mrr_at_1000
value: 37.458000000000006
- type: mrr_at_20
value: 37.055
- type: mrr_at_3
value: 33.762
- type: mrr_at_5
value: 35.176
- type: ndcg_at_1
value: 25.714
- type: ndcg_at_10
value: 31.902
- type: ndcg_at_100
value: 39.534000000000006
- type: ndcg_at_1000
value: 42.424
- type: ndcg_at_20
value: 35.189
- type: ndcg_at_3
value: 26.52
- type: ndcg_at_5
value: 27.833999999999996
- type: precision_at_1
value: 25.714
- type: precision_at_10
value: 9.171
- type: precision_at_100
value: 1.569
- type: precision_at_1000
value: 0.197
- type: precision_at_20
value: 5.671
- type: precision_at_3
value: 18.095
- type: precision_at_5
value: 13.370999999999999
- type: recall_at_1
value: 12.864
- type: recall_at_10
value: 42.522999999999996
- type: recall_at_100
value: 72.653
- type: recall_at_1000
value: 91.49199999999999
- type: recall_at_20
value: 53.452999999999996
- type: recall_at_3
value: 26.007
- type: recall_at_5
value: 31.608999999999998
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (id)
config: id
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 10.563
- type: map_at_10
value: 20.183
- type: map_at_100
value: 22.461000000000002
- type: map_at_1000
value: 22.719
- type: map_at_20
value: 21.342
- type: map_at_3
value: 16.25
- type: map_at_5
value: 18.415
- type: mrr_at_1
value: 23.541999999999998
- type: mrr_at_10
value: 35.272999999999996
- type: mrr_at_100
value: 36.291000000000004
- type: mrr_at_1000
value: 36.336
- type: mrr_at_20
value: 35.894999999999996
- type: mrr_at_3
value: 31.874999999999996
- type: mrr_at_5
value: 33.833
- type: ndcg_at_1
value: 23.541999999999998
- type: ndcg_at_10
value: 28.144000000000002
- type: ndcg_at_100
value: 36.806
- type: ndcg_at_1000
value: 40.888000000000005
- type: ndcg_at_20
value: 31.263
- type: ndcg_at_3
value: 23.796999999999997
- type: ndcg_at_5
value: 25.304
- type: precision_at_1
value: 23.541999999999998
- type: precision_at_10
value: 9.771
- type: precision_at_100
value: 1.994
- type: precision_at_1000
value: 0.281
- type: precision_at_20
value: 6.343999999999999
- type: precision_at_3
value: 17.639
- type: precision_at_5
value: 14.271
- type: recall_at_1
value: 10.563
- type: recall_at_10
value: 35.47
- type: recall_at_100
value: 65.713
- type: recall_at_1000
value: 88.68599999999999
- type: recall_at_20
value: 44.285999999999994
- type: recall_at_3
value: 21.366
- type: recall_at_5
value: 27.598
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (indonesian)
config: indonesian
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 17.149
- type: map_at_10
value: 27.351
- type: map_at_100
value: 28.42
- type: map_at_1000
value: 28.491
- type: map_at_20
value: 28.026
- type: map_at_3
value: 23.918
- type: map_at_5
value: 26.125999999999998
- type: mrr_at_1
value: 18.938
- type: mrr_at_10
value: 28.853
- type: mrr_at_100
value: 29.805999999999997
- type: mrr_at_1000
value: 29.863
- type: mrr_at_20
value: 29.457
- type: mrr_at_3
value: 25.714
- type: mrr_at_5
value: 27.674
- type: ndcg_at_1
value: 18.938
- type: ndcg_at_10
value: 33.195
- type: ndcg_at_100
value: 38.17
- type: ndcg_at_1000
value: 40.109
- type: ndcg_at_20
value: 35.514
- type: ndcg_at_3
value: 26.509
- type: ndcg_at_5
value: 30.325000000000003
- type: precision_at_1
value: 18.938
- type: precision_at_10
value: 5.694
- type: precision_at_100
value: 0.8370000000000001
- type: precision_at_1000
value: 0.10200000000000001
- type: precision_at_20
value: 3.366
- type: precision_at_3
value: 12.142999999999999
- type: precision_at_5
value: 9.385
- type: recall_at_1
value: 17.149
- type: recall_at_10
value: 49.799
- type: recall_at_100
value: 72.256
- type: recall_at_1000
value: 87.354
- type: recall_at_20
value: 58.585
- type: recall_at_3
value: 32.268
- type: recall_at_5
value: 41.275
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (ja)
config: ja
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 10.001999999999999
- type: map_at_10
value: 19.887
- type: map_at_100
value: 22.192
- type: map_at_1000
value: 22.364
- type: map_at_20
value: 21.127000000000002
- type: map_at_3
value: 16.297
- type: map_at_5
value: 18.046
- type: mrr_at_1
value: 17.442
- type: mrr_at_10
value: 29.804000000000002
- type: mrr_at_100
value: 31.025999999999996
- type: mrr_at_1000
value: 31.075999999999997
- type: mrr_at_20
value: 30.570000000000004
- type: mrr_at_3
value: 26.491999999999997
- type: mrr_at_5
value: 28.341
- type: ndcg_at_1
value: 17.442
- type: ndcg_at_10
value: 27.561999999999998
- type: ndcg_at_100
value: 36.714999999999996
- type: ndcg_at_1000
value: 39.564
- type: ndcg_at_20
value: 31.133
- type: ndcg_at_3
value: 21.464
- type: ndcg_at_5
value: 23.775
- type: precision_at_1
value: 17.442
- type: precision_at_10
value: 7.663
- type: precision_at_100
value: 1.566
- type: precision_at_1000
value: 0.197
- type: precision_at_20
value: 5.07
- type: precision_at_3
value: 14.302000000000001
- type: precision_at_5
value: 11.07
- type: recall_at_1
value: 10.001999999999999
- type: recall_at_10
value: 39.982
- type: recall_at_100
value: 76.165
- type: recall_at_1000
value: 94.487
- type: recall_at_20
value: 51.559999999999995
- type: recall_at_3
value: 23.274
- type: recall_at_5
value: 29.842000000000002
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (japanese)
config: japanese
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 6.412
- type: map_at_10
value: 12.008000000000001
- type: map_at_100
value: 13.138
- type: map_at_1000
value: 13.227
- type: map_at_20
value: 12.669
- type: map_at_3
value: 9.738
- type: map_at_5
value: 10.852
- type: mrr_at_1
value: 8.056000000000001
- type: mrr_at_10
value: 14.155999999999999
- type: mrr_at_100
value: 15.232000000000001
- type: mrr_at_1000
value: 15.308
- type: mrr_at_20
value: 14.780999999999999
- type: mrr_at_3
value: 11.806
- type: mrr_at_5
value: 12.992999999999999
- type: ndcg_at_1
value: 8.056000000000001
- type: ndcg_at_10
value: 16.048000000000002
- type: ndcg_at_100
value: 21.833
- type: ndcg_at_1000
value: 24.282
- type: ndcg_at_20
value: 18.398
- type: ndcg_at_3
value: 11.317
- type: ndcg_at_5
value: 13.319
- type: precision_at_1
value: 8.056000000000001
- type: precision_at_10
value: 3.292
- type: precision_at_100
value: 0.664
- type: precision_at_1000
value: 0.091
- type: precision_at_20
value: 2.181
- type: precision_at_3
value: 5.833
- type: precision_at_5
value: 4.583
- type: recall_at_1
value: 6.412
- type: recall_at_10
value: 26.480999999999998
- type: recall_at_100
value: 53.217999999999996
- type: recall_at_1000
value: 71.528
- type: recall_at_20
value: 35.44
- type: recall_at_3
value: 13.866
- type: recall_at_5
value: 18.519
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (ko)
config: ko
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 19.03
- type: map_at_10
value: 30.987
- type: map_at_100
value: 33.248
- type: map_at_1000
value: 33.452
- type: map_at_20
value: 32.298
- type: map_at_3
value: 26.240999999999996
- type: map_at_5
value: 28.899
- type: mrr_at_1
value: 28.638
- type: mrr_at_10
value: 42.082
- type: mrr_at_100
value: 43.034
- type: mrr_at_1000
value: 43.059
- type: mrr_at_20
value: 42.736000000000004
- type: mrr_at_3
value: 38.340999999999994
- type: mrr_at_5
value: 40.548
- type: ndcg_at_1
value: 28.638
- type: ndcg_at_10
value: 39.284
- type: ndcg_at_100
value: 47.157
- type: ndcg_at_1000
value: 49.864000000000004
- type: ndcg_at_20
value: 42.719
- type: ndcg_at_3
value: 32.521
- type: ndcg_at_5
value: 35.57
- type: precision_at_1
value: 28.638
- type: precision_at_10
value: 10.657
- type: precision_at_100
value: 1.8870000000000002
- type: precision_at_1000
value: 0.241
- type: precision_at_20
value: 6.643000000000001
- type: precision_at_3
value: 19.718
- type: precision_at_5
value: 15.493000000000002
- type: recall_at_1
value: 19.03
- type: recall_at_10
value: 51.961999999999996
- type: recall_at_100
value: 79.836
- type: recall_at_1000
value: 95.3
- type: recall_at_20
value: 62.643
- type: recall_at_3
value: 33.1
- type: recall_at_5
value: 41.791
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (korean)
config: korean
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 18.131
- type: map_at_10
value: 23.858999999999998
- type: map_at_100
value: 24.507
- type: map_at_1000
value: 24.582
- type: map_at_20
value: 24.152
- type: map_at_3
value: 22.09
- type: map_at_5
value: 22.971
- type: mrr_at_1
value: 20.19
- type: mrr_at_10
value: 26.249
- type: mrr_at_100
value: 26.871000000000002
- type: mrr_at_1000
value: 26.936
- type: mrr_at_20
value: 26.535999999999998
- type: mrr_at_3
value: 24.465999999999998
- type: mrr_at_5
value: 25.356
- type: ndcg_at_1
value: 20.19
- type: ndcg_at_10
value: 27.700000000000003
- type: ndcg_at_100
value: 31.436999999999998
- type: ndcg_at_1000
value: 33.690999999999995
- type: ndcg_at_20
value: 28.716
- type: ndcg_at_3
value: 23.985
- type: ndcg_at_5
value: 25.539
- type: precision_at_1
value: 20.19
- type: precision_at_10
value: 4.181
- type: precision_at_100
value: 0.637
- type: precision_at_1000
value: 0.08499999999999999
- type: precision_at_20
value: 2.316
- type: precision_at_3
value: 10.135
- type: precision_at_5
value: 6.888
- type: recall_at_1
value: 18.131
- type: recall_at_10
value: 37.49
- type: recall_at_100
value: 55.859
- type: recall_at_1000
value: 73.634
- type: recall_at_20
value: 41.291
- type: recall_at_3
value: 27.276
- type: recall_at_5
value: 30.918
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (ru)
config: ru
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 12.045
- type: map_at_10
value: 21.751
- type: map_at_100
value: 24.517
- type: map_at_1000
value: 24.726
- type: map_at_20
value: 23.142
- type: map_at_3
value: 17.546
- type: map_at_5
value: 19.555
- type: mrr_at_1
value: 22.900000000000002
- type: mrr_at_10
value: 33.946
- type: mrr_at_100
value: 35.199999999999996
- type: mrr_at_1000
value: 35.233
- type: mrr_at_20
value: 34.791
- type: mrr_at_3
value: 30.099999999999998
- type: mrr_at_5
value: 32.18
- type: ndcg_at_1
value: 22.900000000000002
- type: ndcg_at_10
value: 29.587000000000003
- type: ndcg_at_100
value: 39.934999999999995
- type: ndcg_at_1000
value: 42.857
- type: ndcg_at_20
value: 33.531
- type: ndcg_at_3
value: 23.361
- type: ndcg_at_5
value: 25.485999999999997
- type: precision_at_1
value: 22.900000000000002
- type: precision_at_10
value: 9.370000000000001
- type: precision_at_100
value: 2.1
- type: precision_at_1000
value: 0.263
- type: precision_at_20
value: 6.4399999999999995
- type: precision_at_3
value: 15.867
- type: precision_at_5
value: 12.86
- type: recall_at_1
value: 12.045
- type: recall_at_10
value: 39.612
- type: recall_at_100
value: 76.132
- type: recall_at_1000
value: 92.643
- type: recall_at_20
value: 51.13799999999999
- type: recall_at_3
value: 22.408
- type: recall_at_5
value: 28.794999999999998
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (russian)
config: russian
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 12.127
- type: map_at_10
value: 18.472
- type: map_at_100
value: 19.395
- type: map_at_1000
value: 19.477
- type: map_at_20
value: 18.966
- type: map_at_3
value: 16.309
- type: map_at_5
value: 17.53
- type: mrr_at_1
value: 13.568
- type: mrr_at_10
value: 20.04
- type: mrr_at_100
value: 20.928
- type: mrr_at_1000
value: 20.995
- type: mrr_at_20
value: 20.538
- type: mrr_at_3
value: 17.855999999999998
- type: mrr_at_5
value: 19.027
- type: ndcg_at_1
value: 13.568
- type: ndcg_at_10
value: 22.509
- type: ndcg_at_100
value: 27.230999999999998
- type: ndcg_at_1000
value: 29.444
- type: ndcg_at_20
value: 24.277
- type: ndcg_at_3
value: 18.108
- type: ndcg_at_5
value: 20.252
- type: precision_at_1
value: 13.568
- type: precision_at_10
value: 3.769
- type: precision_at_100
value: 0.635
- type: precision_at_1000
value: 0.08499999999999999
- type: precision_at_20
value: 2.271
- type: precision_at_3
value: 8.107000000000001
- type: precision_at_5
value: 5.99
- type: recall_at_1
value: 12.127
- type: recall_at_10
value: 33.617999999999995
- type: recall_at_100
value: 55.662
- type: recall_at_1000
value: 72.697
- type: recall_at_20
value: 40.369
- type: recall_at_3
value: 21.91
- type: recall_at_5
value: 26.985
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (es)
config: es
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 8.068
- type: map_at_10
value: 19.563
- type: map_at_100
value: 24.198
- type: map_at_1000
value: 24.532
- type: map_at_20
value: 21.929000000000002
- type: map_at_3
value: 13.652000000000001
- type: map_at_5
value: 16.226
- type: mrr_at_1
value: 30.092999999999996
- type: mrr_at_10
value: 43.312
- type: mrr_at_100
value: 44.230999999999995
- type: mrr_at_1000
value: 44.252
- type: mrr_at_20
value: 43.891000000000005
- type: mrr_at_3
value: 39.428999999999995
- type: mrr_at_5
value: 41.682
- type: ndcg_at_1
value: 30.092999999999996
- type: ndcg_at_10
value: 30.218
- type: ndcg_at_100
value: 43.716
- type: ndcg_at_1000
value: 47.355000000000004
- type: ndcg_at_20
value: 35.268
- type: ndcg_at_3
value: 27.232
- type: ndcg_at_5
value: 26.874
- type: precision_at_1
value: 30.092999999999996
- type: precision_at_10
value: 15.293000000000001
- type: precision_at_100
value: 3.531
- type: precision_at_1000
value: 0.437
- type: precision_at_20
value: 10.693999999999999
- type: precision_at_3
value: 23.868000000000002
- type: precision_at_5
value: 19.938
- type: recall_at_1
value: 8.068
- type: recall_at_10
value: 35.778999999999996
- type: recall_at_100
value: 76.36699999999999
- type: recall_at_1000
value: 94.38499999999999
- type: recall_at_20
value: 48.014
- type: recall_at_3
value: 17.825
- type: recall_at_5
value: 24.163999999999998
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (sw)
config: sw
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 12.568999999999999
- type: map_at_10
value: 18.478
- type: map_at_100
value: 19.584
- type: map_at_1000
value: 19.714000000000002
- type: map_at_20
value: 19.073999999999998
- type: map_at_3
value: 16.275000000000002
- type: map_at_5
value: 17.53
- type: mrr_at_1
value: 20.124
- type: mrr_at_10
value: 25.951
- type: mrr_at_100
value: 26.812
- type: mrr_at_1000
value: 26.876
- type: mrr_at_20
value: 26.41
- type: mrr_at_3
value: 23.893
- type: mrr_at_5
value: 25.055
- type: ndcg_at_1
value: 20.124
- type: ndcg_at_10
value: 23.064999999999998
- type: ndcg_at_100
value: 28.193
- type: ndcg_at_1000
value: 31.313000000000002
- type: ndcg_at_20
value: 24.921
- type: ndcg_at_3
value: 19.436
- type: ndcg_at_5
value: 20.990000000000002
- type: precision_at_1
value: 20.124
- type: precision_at_10
value: 5.021
- type: precision_at_100
value: 0.927
- type: precision_at_1000
value: 0.13699999999999998
- type: precision_at_20
value: 3.1329999999999996
- type: precision_at_3
value: 10.719
- type: precision_at_5
value: 8.091
- type: recall_at_1
value: 12.568999999999999
- type: recall_at_10
value: 29.727999999999998
- type: recall_at_100
value: 50.70099999999999
- type: recall_at_1000
value: 71.856
- type: recall_at_20
value: 35.831
- type: recall_at_3
value: 19.758
- type: recall_at_5
value: 24.035999999999998
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (swahili)
config: swahili
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 15.597
- type: map_at_10
value: 23.596
- type: map_at_100
value: 24.32
- type: map_at_1000
value: 24.377
- type: map_at_20
value: 23.913999999999998
- type: map_at_3
value: 21.484
- type: map_at_5
value: 22.647000000000002
- type: mrr_at_1
value: 16.866
- type: mrr_at_10
value: 24.633
- type: mrr_at_100
value: 25.291000000000004
- type: mrr_at_1000
value: 25.346999999999998
- type: mrr_at_20
value: 24.907
- type: mrr_at_3
value: 22.587
- type: mrr_at_5
value: 23.677
- type: ndcg_at_1
value: 16.866
- type: ndcg_at_10
value: 27.749000000000002
- type: ndcg_at_100
value: 31.769
- type: ndcg_at_1000
value: 33.61
- type: ndcg_at_20
value: 28.849000000000004
- type: ndcg_at_3
value: 23.462
- type: ndcg_at_5
value: 25.491000000000003
- type: precision_at_1
value: 16.866
- type: precision_at_10
value: 4.343
- type: precision_at_100
value: 0.649
- type: precision_at_1000
value: 0.08099999999999999
- type: precision_at_20
value: 2.41
- type: precision_at_3
value: 10.149
- type: precision_at_5
value: 7.134
- type: recall_at_1
value: 15.597
- type: recall_at_10
value: 39.652
- type: recall_at_100
value: 59.477999999999994
- type: recall_at_1000
value: 74.35300000000001
- type: recall_at_20
value: 43.905
- type: recall_at_3
value: 28.109
- type: recall_at_5
value: 32.934999999999995
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (te)
config: te
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 21.236
- type: map_at_10
value: 29.841
- type: map_at_100
value: 30.729
- type: map_at_1000
value: 30.817
- type: map_at_20
value: 30.331999999999997
- type: map_at_3
value: 27.154
- type: map_at_5
value: 28.719
- type: mrr_at_1
value: 21.618000000000002
- type: mrr_at_10
value: 30.192999999999998
- type: mrr_at_100
value: 31.067
- type: mrr_at_1000
value: 31.148999999999997
- type: mrr_at_20
value: 30.668
- type: mrr_at_3
value: 27.576
- type: mrr_at_5
value: 29.110000000000003
- type: ndcg_at_1
value: 21.618000000000002
- type: ndcg_at_10
value: 34.633
- type: ndcg_at_100
value: 39.425
- type: ndcg_at_1000
value: 41.762
- type: ndcg_at_20
value: 36.428
- type: ndcg_at_3
value: 29.189
- type: ndcg_at_5
value: 31.995
- type: precision_at_1
value: 21.618000000000002
- type: precision_at_10
value: 5.1209999999999996
- type: precision_at_100
value: 0.751
- type: precision_at_1000
value: 0.094
- type: precision_at_20
value: 2.923
- type: precision_at_3
value: 11.876000000000001
- type: precision_at_5
value: 8.551
- type: recall_at_1
value: 21.236
- type: recall_at_10
value: 49.436
- type: recall_at_100
value: 72.94699999999999
- type: recall_at_1000
value: 91.405
- type: recall_at_20
value: 56.562
- type: recall_at_3
value: 34.682
- type: recall_at_5
value: 41.445
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (telugu)
config: telugu
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 16.872999999999998
- type: map_at_10
value: 26.543
- type: map_at_100
value: 27.3
- type: map_at_1000
value: 27.389999999999997
- type: map_at_20
value: 26.883000000000003
- type: map_at_3
value: 23.735999999999997
- type: map_at_5
value: 25.113999999999997
- type: mrr_at_1
value: 17.492
- type: mrr_at_10
value: 27.233
- type: mrr_at_100
value: 27.950999999999997
- type: mrr_at_1000
value: 28.037
- type: mrr_at_20
value: 27.556000000000004
- type: mrr_at_3
value: 24.458
- type: mrr_at_5
value: 25.874999999999996
- type: ndcg_at_1
value: 17.492
- type: ndcg_at_10
value: 32.013999999999996
- type: ndcg_at_100
value: 36.299
- type: ndcg_at_1000
value: 38.868
- type: ndcg_at_20
value: 33.238
- type: ndcg_at_3
value: 26.133
- type: ndcg_at_5
value: 28.632999999999996
- type: precision_at_1
value: 17.492
- type: precision_at_10
value: 5.108
- type: precision_at_100
value: 0.735
- type: precision_at_1000
value: 0.095
- type: precision_at_20
value: 2.81
- type: precision_at_3
value: 11.3
- type: precision_at_5
value: 8.019
- type: recall_at_1
value: 16.872999999999998
- type: recall_at_10
value: 48.684
- type: recall_at_100
value: 70.04599999999999
- type: recall_at_1000
value: 90.635
- type: recall_at_20
value: 53.483000000000004
- type: recall_at_3
value: 32.353
- type: recall_at_5
value: 38.39
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (th)
config: th
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 19.205
- type: map_at_10
value: 32.704
- type: map_at_100
value: 34.519
- type: map_at_1000
value: 34.634
- type: map_at_20
value: 33.762
- type: map_at_3
value: 27.992
- type: map_at_5
value: 30.354999999999997
- type: mrr_at_1
value: 27.558
- type: mrr_at_10
value: 40.933
- type: mrr_at_100
value: 41.916
- type: mrr_at_1000
value: 41.946
- type: mrr_at_20
value: 41.54
- type: mrr_at_3
value: 37.312
- type: mrr_at_5
value: 39.393
- type: ndcg_at_1
value: 27.558
- type: ndcg_at_10
value: 41.038999999999994
- type: ndcg_at_100
value: 48.076
- type: ndcg_at_1000
value: 49.946
- type: ndcg_at_20
value: 44.03
- type: ndcg_at_3
value: 33.013999999999996
- type: ndcg_at_5
value: 36.519
- type: precision_at_1
value: 27.558
- type: precision_at_10
value: 9.523
- type: precision_at_100
value: 1.52
- type: precision_at_1000
value: 0.178
- type: precision_at_20
value: 5.778
- type: precision_at_3
value: 18.963
- type: precision_at_5
value: 14.379
- type: recall_at_1
value: 19.205
- type: recall_at_10
value: 56.94200000000001
- type: recall_at_100
value: 84.937
- type: recall_at_1000
value: 96.86399999999999
- type: recall_at_20
value: 66.719
- type: recall_at_3
value: 36.536
- type: recall_at_5
value: 44.854
- task:
type: retrieval
dataset:
type: mteb/mrtydi
name: MTEB MrTydiRetrieval (thai)
config: thai
split: test
revision: fc24a3ce8f09746410daee3d5cd823ff7a0675b7
metrics:
- type: map_at_1
value: 17.913
- type: map_at_10
value: 29.006999999999998
- type: map_at_100
value: 30.184
- type: map_at_1000
value: 30.249
- type: map_at_20
value: 29.799999999999997
- type: map_at_3
value: 25.14
- type: map_at_5
value: 27.172
- type: mrr_at_1
value: 19.664
- type: mrr_at_10
value: 30.891000000000002
- type: mrr_at_100
value: 31.896
- type: mrr_at_1000
value: 31.951
- type: mrr_at_20
value: 31.566
- type: mrr_at_3
value: 27.128999999999998
- type: mrr_at_5
value: 29.255
- type: ndcg_at_1
value: 19.664
- type: ndcg_at_10
value: 35.674
- type: ndcg_at_100
value: 40.969
- type: ndcg_at_1000
value: 42.698
- type: ndcg_at_20
value: 38.345
- type: ndcg_at_3
value: 27.826
- type: ndcg_at_5
value: 31.493
- type: precision_at_1
value: 19.664
- type: precision_at_10
value: 6.143
- type: precision_at_100
value: 0.8909999999999999
- type: precision_at_1000
value: 0.105
- type: precision_at_20
value: 3.66
- type: precision_at_3
value: 12.353
- type: precision_at_5
value: 9.411999999999999
- type: recall_at_1
value: 17.913
- type: recall_at_10
value: 54.705999999999996
- type: recall_at_100
value: 78.557
- type: recall_at_1000
value: 91.835
- type: recall_at_20
value: 64.846
- type: recall_at_3
value: 33.669
- type: recall_at_5
value: 42.297000000000004
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (yo)
config: yo
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 7.002999999999999
- type: map_at_10
value: 9.567
- type: map_at_100
value: 10.126
- type: map_at_1000
value: 10.209
- type: map_at_20
value: 9.944
- type: map_at_3
value: 8.929
- type: map_at_5
value: 8.929
- type: mrr_at_1
value: 7.563000000000001
- type: mrr_at_10
value: 10.503
- type: mrr_at_100
value: 11.151
- type: mrr_at_1000
value: 11.232000000000001
- type: mrr_at_20
value: 10.965
- type: mrr_at_3
value: 9.804
- type: mrr_at_5
value: 9.804
- type: ndcg_at_1
value: 7.563000000000001
- type: ndcg_at_10
value: 11.304
- type: ndcg_at_100
value: 14.526
- type: ndcg_at_1000
value: 17.253
- type: ndcg_at_20
value: 12.766
- type: ndcg_at_3
value: 9.797
- type: ndcg_at_5
value: 9.754999999999999
- type: precision_at_1
value: 7.563000000000001
- type: precision_at_10
value: 1.765
- type: precision_at_100
value: 0.378
- type: precision_at_1000
value: 0.064
- type: precision_at_20
value: 1.261
- type: precision_at_3
value: 4.202
- type: precision_at_5
value: 2.521
- type: recall_at_1
value: 7.002999999999999
- type: recall_at_10
value: 16.036
- type: recall_at_100
value: 31.302999999999997
- type: recall_at_1000
value: 53.010999999999996
- type: recall_at_20
value: 21.429000000000002
- type: recall_at_3
value: 11.415000000000001
- type: recall_at_5
value: 11.415000000000001
- task:
type: retrieval
dataset:
type: mteb/miracl-hard-negatives
name: MTEB MIRACLRetrievalHardNegatives (zh)
config: zh
split: dev
revision: 95c8db7d4a6e9c1d8a60601afd63d553ae20a2eb
metrics:
- type: map_at_1
value: 8.341999999999999
- type: map_at_10
value: 18.393
- type: map_at_100
value: 21.796
- type: map_at_1000
value: 21.932
- type: map_at_20
value: 20.238999999999997
- type: map_at_3
value: 13.386999999999999
- type: map_at_5
value: 15.568000000000001
- type: mrr_at_1
value: 15.522
- type: mrr_at_10
value: 28.386
- type: mrr_at_100
value: 29.906
- type: mrr_at_1000
value: 29.925
- type: mrr_at_20
value: 29.413
- type: mrr_at_3
value: 23.707
- type: mrr_at_5
value: 26.073
- type: ndcg_at_1
value: 15.522
- type: ndcg_at_10
value: 26.904
- type: ndcg_at_100
value: 39.336
- type: ndcg_at_1000
value: 41.244
- type: ndcg_at_20
value: 32.04
- type: ndcg_at_3
value: 18.34
- type: ndcg_at_5
value: 21.29
- type: precision_at_1
value: 15.522
- type: precision_at_10
value: 9.084
- type: precision_at_100
value: 2.1149999999999998
- type: precision_at_1000
value: 0.244
- type: precision_at_20
value: 6.5009999999999994
- type: precision_at_3
value: 13.232
- type: precision_at_5
value: 11.552
- type: recall_at_1
value: 8.341999999999999
- type: recall_at_10
value: 40.553
- type: recall_at_100
value: 86.33
- type: recall_at_1000
value: 97.24199999999999
- type: recall_at_20
value: 56.589999999999996
- type: recall_at_3
value: 18.82
- type: recall_at_5
value: 26.304
base_model: FacebookAI/xlm-roberta-base
widget:
- source_sentence: how to get rid of an iron mark
sentences:
- >-
Quick Answer. A good remedy for removing shiny iron scorch marks from
fabric is to use hydrogen peroxide with ammonia. Other options for
removing shiny scorch marks include laundry detergent, bleach or
vinegar, but it depends on how quickly the scorch is remedied. Keep
Learning.
- >-
Largely due to declining sales, in 2006, Tommy Hilfiger sold his company
for $1.6 billion, or $16.80 a share, to Apax Partners, a private
investment company. In March 2010, Phillips-Van Heusen, owner of Calvin
Klein, bought the Tommy Hilfiger Corporation for $3 billion.
- >-
You need to heat continuous until it turns to a paste. Use this simple
mixture by rubbing it right onto the soleplate. Now, make sure that the
iron is unplugged before cleaning it. After rubbing the mixture, with
the help of a nice, clean cloth wipe the unsightly scorch marks off your
iron. 5 people found this useful.
- source_sentence: how much does ipl facial cost
sentences:
- "Prices usually vary according to the ipl treatment size. The average cost for a FotoFacial/IPL is $350 - $600 each treatment, depending on the body part.A consultation with a fotofacial specialist and the number of treatments needed will determine your ipl treatment cost.s discussed above, IPL does not damage the skin surface, unlike dermabrasion and laser resurfacing. Therefore, there is virtually no recovery time.â\x80\x9D Treatments take approximately 30-45 minutes. Patients can apply makeup before leaving the office and return to work the same day."
- "â\x80\x94N.I., South Kingstown, Rhode IslandGenerally, the oven temperature will not need to be adjusted when baking mini or jumbo muffins, but the baking time will most likely need to be altered. Mini muffins will take anywhere from 10 to 15 minutes while jumbo muffins will bake from 20 to 40 minutes.Check jumbo muffins for doneness after 20 minutes, then every 5 to 10 minutes. Keep in mind that the baking time will vary according to the recipe.The variation is due to the oven temperature and the amount of batter in each muffin cup.heck jumbo muffins for doneness after 20 minutes, then every 5 to 10 minutes. Keep in mind that the baking time will vary according to the recipe. The variation is due to the oven temperature and the amount of batter in each muffin cup."
- >-
Prices usually vary according to the ipl treatment size. The average
cost for a FotoFacial/IPL is $350 - $600 each treatment, depending on
the body part.A consultation with a fotofacial specialist and the number
of treatments needed will determine your ipl treatment cost.PL, which
stands for intensed pulsed light, is non-ablative meaning that is does
not damage the surface of the skin. The intense light is delivered to
the deeper parts of the skin (dermis) and leaves the superficial aspect
of the skin (epidermis) untouched.
- source_sentence: who voiced scooby doo
sentences:
- >-
Don Messick originated the voice of Scooby-Doo, and was the voice of the
character for over 25 years until his retirement from voice acting in
1996 (he subsequently passed away the following year).rank Welker's
Scooby-Doo voice is pretty much identical to the voice he used for Brain
on Inspector Gadget.. Hadley Kay and Neil Fanning are the worst, IMO...
- >-
because als symptoms include fatigue muscle weakness and muscle twitches
early on it can look like other very treatable illnesses one that
commonly comes up is lyme disease an infectious disease resulting from a
tick bite unlike als lyme is usually treatable with antibiotics lyme
disease does not cause als and generally in a diagnostic workup a
neurologist can easily separate als from lyme infections either
clinically or with testing
- "Curse Of The Lake Monster while Frank Welker voices him. 1 Don Messick (1969â\x80\x931996) 2 Hadley Kay (Johnny Bravo) 3 Scott Innes (1998â\x80\x932001) Frank Welker (2002â\x80\x93present plus Scooby- 1 Doo! Neil Fanning (2002 and 2004 live-action films) Dave Coulier(2005 in Robot 1 Chicken) In Denmark, Scooby Doo is voiced by Lars Thiesgaard."
- source_sentence: track lighting that can be mounted on wall
sentences:
- >-
Madison is one of 14 Community Plan areas in the Metro
Nashville-Davidson County area for which zoning and land use planning is
done. The 2015-updated Community Plan for Madison, an 89-page document
adopted by the Metropolitan Planning Commission, was updated in 2015 as
part of NashvilleNext's long-term planning.
- >-
This three-light plug-in LED track kit can be surface-mounted anywhere
in a room as the power feed cord eliminates the need for a ...junction
box. Quick and easy to install, it features three 12 watt LED bullet
heads that pivot in a cradle and produce a spotlight beam of
energy-saving light.
- >-
This white finish, three-light LED track kit can be surface-mounted or
suspended from the ceiling with pendant lighting accessorie...s. A power
feed cord and plug lets you install it easily without the need for a
junction box. Arrange the three,...
- source_sentence: Most Common Apple Varieties
sentences:
- "Well, rest easy, because this condensed list of the 18 most popular apple varieties breaks down the information every apple eater should knowâ\x80\x94how to cook them, best recipes, and when they are in season. Red Delicious: A popular eating apple that looks just how we all imagine an apple should."
- >-
Here's a look at the top 50 draft-eligible prospects for next year, led
by quarterback Connor Cook, defensive lineman Joey Bosa, cornerback
Vernon Hargreaves, and offensive tackle Ronnie Stanley and Laremy
Tunsil.
- >-
The most popular apple varieties are Cortland, Red Delicious, Golden
Delicious, Empire, Fuji, Gala, Ida Red, Macoun, McIntosh, Northern Spy,
and Winesap. Olwen Woodier also offers descriptions for an additional 20
varieties of apples in this very useful and informative cookbook.
Cortland.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
license: mit
SentenceTransformer based on FacebookAI/xlm-roberta-base
This is a sentence-transformers model finetuned from FacebookAI/xlm-roberta-base. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
Model Details
Model Description
- Model Type: Sentence Transformer
- Base model: FacebookAI/xlm-roberta-base
- Maximum Sequence Length: 512 tokens
- Output Dimensionality: 768 dimensions
- Similarity Function: Cosine Similarity
- Training Dataset: MS MARCO Hard Negatives (mined by a MiniLM cross-encoder)
- License: MIT
Model Sources
- Documentation: Sentence Transformers Documentation
- Repository: Sentence Transformers on GitHub
- Hugging Face: Sentence Transformers on Hugging Face
Full Model Architecture
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)
Usage
Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
pip install -U sentence-transformers
Then you can load this model and run inference.
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("PaDaS-Lab/xlm-roberta-base-msmarco")
# Run inference
sentences = [
'Most Common Apple Varieties',
'The most popular apple varieties are Cortland, Red Delicious, Golden Delicious, Empire, Fuji, Gala, Ida Red, Macoun, McIntosh, Northern Spy, and Winesap. Olwen Woodier also offers descriptions for an additional 20 varieties of apples in this very useful and informative cookbook. Cortland.',
'Well, rest easy, because this condensed list of the 18 most popular apple varieties breaks down the information every apple eater should knowâ\x80\x94how to cook them, best recipes, and when they are in season. Red Delicious: A popular eating apple that looks just how we all imagine an apple should.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
Training Details
Training Dataset
Unnamed Dataset
- Size: 502,912 training samples
- Columns:
sentence_0,sentence_1,sentence_2, andlabel - Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 sentence_2 label type string string string float details - min: 4 tokens
- mean: 9.88 tokens
- max: 59 tokens
- min: 19 tokens
- mean: 88.5 tokens
- max: 232 tokens
- min: 17 tokens
- mean: 87.87 tokens
- max: 282 tokens
- min: -16.56
- mean: 0.96
- max: 20.84
- Samples:
sentence_0 sentence_1 sentence_2 label how long are bank issued checks good forYour mom is correct....most checks are good for anywhere between 180 days up to 1 year. Sorry, but you probably won't be able to cash those checks, although it never hurts to check with your bank on the issue. DH · 9 years ago.Non-local personal and business checks. If the check is from a bank in a different federal reserve district than the depositing bank, it can be held for 5 business days under normal circumstances. Exceptions for new customers during the first 30 days. Banks are not required to give next day ability on the first $100 of deposits, and both local and non-local personal and business checks can be held for a maximum of 11 business days.2.652659893035888711:11 meaning11-11-11 11:11:11 example. 11-11 11:11 example. Numerologists believe that events linked to the time 11:11 appear more often than can be explained by chance or coincidence. This belief is related to the concept of synchronicity. Some authors claim that seeing 11:11 on a clock is an auspicious sign.Sometimes it's difficult to describe what seeing the 11:11 means, because it is a personal experience for everyone. If you feel you are having these experiences for a reason, then it might be that only you will know what these number prompts and wake-up calls mean.-1.3284940719604492did someone from pawn stars dieDid someone from pawn stars on history channel die? kgb answers » Arts & Entertainment » Actors and Actresses » Did someone from pawn stars on history channel die? None from the actors & cast of Pawn Stars died. There was a rumor that Leonard Shaffer, a coin expert, died but it is not true. He is alive & well. Tags: pawn stars, lists of actors.Austin Russell, also known as Chumlee, star of History's reality series Pawn Stars, has died from an apparent heart attack, sources confirm to eBuzzd.1.7131614685058594 - Loss:
MarginMSELoss
Training Hyperparameters
Non-Default Hyperparameters
per_device_train_batch_size: 64per_device_eval_batch_size: 64num_train_epochs: 30fp16: Truemulti_dataset_batch_sampler: round_robin
All Hyperparameters
Click to expand
overwrite_output_dir: Falsedo_predict: Falseeval_strategy: noprediction_loss_only: Trueper_device_train_batch_size: 64per_device_eval_batch_size: 64per_gpu_train_batch_size: Noneper_gpu_eval_batch_size: Nonegradient_accumulation_steps: 1eval_accumulation_steps: Nonetorch_empty_cache_steps: Nonelearning_rate: 5e-05weight_decay: 0.0adam_beta1: 0.9adam_beta2: 0.999adam_epsilon: 1e-08max_grad_norm: 1num_train_epochs: 30max_steps: -1lr_scheduler_type: linearlr_scheduler_kwargs: {}warmup_ratio: 0.0warmup_steps: 0log_level: passivelog_level_replica: warninglog_on_each_node: Truelogging_nan_inf_filter: Truesave_safetensors: Truesave_on_each_node: Falsesave_only_model: Falserestore_callback_states_from_checkpoint: Falseno_cuda: Falseuse_cpu: Falseuse_mps_device: Falseseed: 42data_seed: Nonejit_mode_eval: Falseuse_ipex: Falsebf16: Falsefp16: Truefp16_opt_level: O1half_precision_backend: autobf16_full_eval: Falsefp16_full_eval: Falsetf32: Nonelocal_rank: 0ddp_backend: Nonetpu_num_cores: Nonetpu_metrics_debug: Falsedebug: []dataloader_drop_last: Falsedataloader_num_workers: 0dataloader_prefetch_factor: Nonepast_index: -1disable_tqdm: Falseremove_unused_columns: Truelabel_names: Noneload_best_model_at_end: Falseignore_data_skip: Falsefsdp: []fsdp_min_num_params: 0fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap: Noneaccelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed: Nonelabel_smoothing_factor: 0.0optim: adamw_torchoptim_args: Noneadafactor: Falsegroup_by_length: Falselength_column_name: lengthddp_find_unused_parameters: Noneddp_bucket_cap_mb: Noneddp_broadcast_buffers: Falsedataloader_pin_memory: Truedataloader_persistent_workers: Falseskip_memory_metrics: Trueuse_legacy_prediction_loop: Falsepush_to_hub: Falseresume_from_checkpoint: Nonehub_model_id: Nonehub_strategy: every_savehub_private_repo: Nonehub_always_push: Falsegradient_checkpointing: Falsegradient_checkpointing_kwargs: Noneinclude_inputs_for_metrics: Falseinclude_for_metrics: []eval_do_concat_batches: Truefp16_backend: autopush_to_hub_model_id: Nonepush_to_hub_organization: Nonemp_parameters:auto_find_batch_size: Falsefull_determinism: Falsetorchdynamo: Noneray_scope: lastddp_timeout: 1800torch_compile: Falsetorch_compile_backend: Nonetorch_compile_mode: Nonedispatch_batches: Nonesplit_batches: Noneinclude_tokens_per_second: Falseinclude_num_input_tokens_seen: Falseneftune_noise_alpha: Noneoptim_target_modules: Nonebatch_eval_metrics: Falseeval_on_start: Falseuse_liger_kernel: Falseeval_use_gather_object: Falseaverage_tokens_across_devices: Falseprompts: Nonebatch_sampler: batch_samplermulti_dataset_batch_sampler: round_robin
Training Logs
Click to expand
| Epoch | Step | Training Loss |
|---|---|---|
| 0.0636 | 500 | 92.5416 |
| 0.1273 | 1000 | 20.6659 |
| 0.1909 | 1500 | 14.7631 |
| 0.2545 | 2000 | 14.3025 |
| 0.3181 | 2500 | 13.5257 |
| 0.3818 | 3000 | 12.8666 |
| 0.4454 | 3500 | 12.397 |
| 0.5090 | 4000 | 12.2718 |
| 0.5727 | 4500 | 11.539 |
| 0.6363 | 5000 | 11.1145 |
| 0.6999 | 5500 | 11.1232 |
| 0.7636 | 6000 | 10.6021 |
| 0.8272 | 6500 | 10.4115 |
| 0.8908 | 7000 | 10.4529 |
| 0.9544 | 7500 | 10.1329 |
| 1.0181 | 8000 | 10.1367 |
| 1.0817 | 8500 | 9.5914 |
| 1.1453 | 9000 | 9.2799 |
| 1.2090 | 9500 | 9.266 |
| 1.2726 | 10000 | 9.1661 |
| 1.3362 | 10500 | 8.954 |
| 1.3998 | 11000 | 8.9562 |
| 1.4635 | 11500 | 9.4717 |
| 1.5271 | 12000 | 8.6758 |
| 1.5907 | 12500 | 8.87 |
| 1.6544 | 13000 | 8.5826 |
| 1.7180 | 13500 | 8.4827 |
| 1.7816 | 14000 | 8.5306 |
| 1.8453 | 14500 | 8.182 |
| 1.9089 | 15000 | 8.3592 |
| 1.9725 | 15500 | 8.3879 |
| 2.0361 | 16000 | 7.4399 |
| 2.0998 | 16500 | 7.0406 |
| 2.1634 | 17000 | 6.89 |
| 2.2270 | 17500 | 6.8651 |
| 2.2907 | 18000 | 6.8461 |
| 2.3543 | 18500 | 6.7663 |
| 2.4179 | 19000 | 6.9313 |
| 2.4815 | 19500 | 6.9688 |
| 2.5452 | 20000 | 6.7821 |
| 2.6088 | 20500 | 6.9468 |
| 2.6724 | 21000 | 6.731 |
| 2.7361 | 21500 | 6.649 |
| 2.7997 | 22000 | 6.7055 |
| 2.8633 | 22500 | 6.7744 |
| 2.9270 | 23000 | 6.9481 |
| 2.9906 | 23500 | 6.5967 |
| 3.0542 | 24000 | 5.7351 |
| 3.1178 | 24500 | 5.4125 |
| 3.1815 | 25000 | 5.4095 |
| 3.2451 | 25500 | 5.4253 |
| 3.3087 | 26000 | 5.3774 |
| 3.3724 | 26500 | 5.5277 |
| 3.4360 | 27000 | 5.4516 |
| 3.4996 | 27500 | 5.322 |
| 3.5632 | 28000 | 5.5531 |
| 3.6269 | 28500 | 5.5238 |
| 3.6905 | 29000 | 5.5992 |
| 3.7541 | 29500 | 5.5351 |
| 3.8178 | 30000 | 5.3985 |
| 3.8814 | 30500 | 5.4313 |
| 3.9450 | 31000 | 5.4173 |
| 4.0087 | 31500 | 5.2333 |
| 4.0723 | 32000 | 4.3352 |
| 4.1359 | 32500 | 4.3442 |
| 4.1995 | 33000 | 4.3288 |
| 4.2632 | 33500 | 4.367 |
| 4.3268 | 34000 | 4.4607 |
| 4.3904 | 34500 | 4.4461 |
| 4.4541 | 35000 | 4.6218 |
| 4.5177 | 35500 | 4.4249 |
| 4.5813 | 36000 | 4.4129 |
| 4.6449 | 36500 | 4.4065 |
| 4.7086 | 37000 | 4.5452 |
| 4.7722 | 37500 | 4.5411 |
| 4.8358 | 38000 | 4.5423 |
| 4.8995 | 38500 | 4.4942 |
| 4.9631 | 39000 | 4.5332 |
| 5.0267 | 39500 | 4.0759 |
| 5.0904 | 40000 | 3.6274 |
| 5.1540 | 40500 | 3.6795 |
| 5.2176 | 41000 | 3.6741 |
| 5.2812 | 41500 | 3.7396 |
| 5.3449 | 42000 | 3.6839 |
| 5.4085 | 42500 | 3.732 |
| 5.4721 | 43000 | 3.6557 |
| 5.5358 | 43500 | 3.6925 |
| 5.5994 | 44000 | 3.7149 |
| 5.6630 | 44500 | 3.6744 |
| 5.7266 | 45000 | 3.7669 |
| 5.7903 | 45500 | 3.651 |
| 5.8539 | 46000 | 3.721 |
| 5.9175 | 46500 | 3.7012 |
| 5.9812 | 47000 | 3.7294 |
| 6.0448 | 47500 | 3.2432 |
| 6.1084 | 48000 | 3.0295 |
| 6.1721 | 48500 | 3.0364 |
| 6.2357 | 49000 | 3.0687 |
| 6.2993 | 49500 | 3.064 |
| 6.3629 | 50000 | 3.112 |
| 6.4266 | 50500 | 3.1438 |
| 6.4902 | 51000 | 3.0733 |
| 6.5538 | 51500 | 3.1719 |
| 6.6175 | 52000 | 3.1355 |
| 6.6811 | 52500 | 3.1612 |
| 6.7447 | 53000 | 3.1938 |
| 6.8083 | 53500 | 3.1375 |
| 6.8720 | 54000 | 3.1969 |
| 6.9356 | 54500 | 3.2214 |
| 6.9992 | 55000 | 3.1364 |
| 7.0629 | 55500 | 2.63 |
| 7.1265 | 56000 | 2.5451 |
| 7.1901 | 56500 | 2.644 |
| 7.2538 | 57000 | 2.6482 |
| 7.3174 | 57500 | 2.6017 |
| 7.3810 | 58000 | 2.6626 |
| 7.4446 | 58500 | 2.6698 |
| 7.5083 | 59000 | 2.6595 |
| 7.5719 | 59500 | 2.6683 |
| 7.6355 | 60000 | 2.7187 |
| 7.6992 | 60500 | 2.6213 |
| 7.7628 | 61000 | 2.7119 |
| 7.8264 | 61500 | 2.739 |
| 7.8900 | 62000 | 2.686 |
| 7.9537 | 62500 | 2.7295 |
| 8.0173 | 63000 | 2.6062 |
| 8.0809 | 63500 | 2.2272 |
| 8.1446 | 64000 | 2.2692 |
| 8.2082 | 64500 | 2.3135 |
| 8.2718 | 65000 | 2.2546 |
| 8.3355 | 65500 | 2.2882 |
| 8.3991 | 66000 | 2.2749 |
| 8.4627 | 66500 | 2.363 |
| 8.5263 | 67000 | 2.2923 |
| 8.5900 | 67500 | 2.3275 |
| 8.6536 | 68000 | 2.3738 |
| 8.7172 | 68500 | 2.3416 |
| 8.7809 | 69000 | 2.3851 |
| 8.8445 | 69500 | 2.3356 |
| 8.9081 | 70000 | 2.3598 |
| 8.9717 | 70500 | 2.4272 |
| 9.0354 | 71000 | 2.141 |
| 9.0990 | 71500 | 2.001 |
| 9.1626 | 72000 | 2.014 |
| 9.2263 | 72500 | 1.9826 |
| 9.2899 | 73000 | 1.995 |
| 9.3535 | 73500 | 2.0097 |
| 9.4172 | 74000 | 2.0412 |
| 9.4808 | 74500 | 2.0144 |
| 9.5444 | 75000 | 2.0653 |
| 9.6080 | 75500 | 2.022 |
| 9.6717 | 76000 | 2.0327 |
| 9.7353 | 76500 | 2.0596 |
| 9.7989 | 77000 | 2.0761 |
| 9.8626 | 77500 | 2.1245 |
| 9.9262 | 78000 | 2.1062 |
| 9.9898 | 78500 | 2.1186 |
| 10.0534 | 79000 | 1.8283 |
| 10.1171 | 79500 | 1.7627 |
| 10.1807 | 80000 | 1.7775 |
| 10.2443 | 80500 | 1.7865 |
| 10.3080 | 81000 | 1.8018 |
| 10.3716 | 81500 | 1.7851 |
| 10.4352 | 82000 | 1.8085 |
| 10.4989 | 82500 | 1.8293 |
| 10.5625 | 83000 | 1.8549 |
| 10.6261 | 83500 | 1.8531 |
| 10.6897 | 84000 | 1.8538 |
| 10.7534 | 84500 | 1.8814 |
| 10.8170 | 85000 | 1.8576 |
| 10.8806 | 85500 | 1.8516 |
| 10.9443 | 86000 | 1.8555 |
| 11.0079 | 86500 | 1.8631 |
| 11.0715 | 87000 | 1.6189 |
| 11.1351 | 87500 | 1.6143 |
| 11.1988 | 88000 | 1.6246 |
| 11.2624 | 88500 | 1.5997 |
| 11.3260 | 89000 | 1.646 |
| 11.3897 | 89500 | 1.6323 |
| 11.4533 | 90000 | 1.6623 |
| 11.5169 | 90500 | 1.6544 |
| 11.5806 | 91000 | 1.6671 |
| 11.6442 | 91500 | 1.6742 |
| 11.7078 | 92000 | 1.6409 |
| 11.7714 | 92500 | 1.6504 |
| 11.8351 | 93000 | 1.6791 |
| 11.8987 | 93500 | 1.6923 |
| 11.9623 | 94000 | 1.697 |
| 12.0260 | 94500 | 1.6136 |
| 12.0896 | 95000 | 1.4437 |
| 12.1532 | 95500 | 1.49 |
| 12.2168 | 96000 | 1.4567 |
| 12.2805 | 96500 | 1.5007 |
| 12.3441 | 97000 | 1.4826 |
| 12.4077 | 97500 | 1.4668 |
| 12.4714 | 98000 | 1.5009 |
| 12.5350 | 98500 | 1.5008 |
| 12.5986 | 99000 | 1.5336 |
| 12.6623 | 99500 | 1.5057 |
| 12.7259 | 100000 | 1.5081 |
| 12.7895 | 100500 | 1.5402 |
| 12.8531 | 101000 | 1.5519 |
| 12.9168 | 101500 | 1.5171 |
| 12.9804 | 102000 | 1.5249 |
| 13.0440 | 102500 | 1.4117 |
| 13.1077 | 103000 | 1.3524 |
| 13.1713 | 103500 | 1.3564 |
| 13.2349 | 104000 | 1.3483 |
| 13.2985 | 104500 | 1.386 |
| 13.3622 | 105000 | 1.3723 |
| 13.4258 | 105500 | 1.3933 |
| 13.4894 | 106000 | 1.3672 |
| 13.5531 | 106500 | 1.3796 |
| 13.6167 | 107000 | 1.3637 |
| 13.6803 | 107500 | 1.4061 |
| 13.7440 | 108000 | 1.3897 |
| 13.8076 | 108500 | 1.4342 |
| 13.8712 | 109000 | 1.3821 |
| 13.9348 | 109500 | 1.411 |
| 13.9985 | 110000 | 1.4214 |
| 14.0621 | 110500 | 1.2551 |
| 14.1257 | 111000 | 1.2366 |
| 14.1894 | 111500 | 1.2553 |
| 14.2530 | 112000 | 1.2553 |
| 14.3166 | 112500 | 1.2624 |
| 14.3802 | 113000 | 1.2771 |
| 14.4439 | 113500 | 1.2744 |
| 14.5075 | 114000 | 1.2616 |
| 14.5711 | 114500 | 1.2744 |
| 14.6348 | 115000 | 1.2705 |
| 14.6984 | 115500 | 1.3005 |
| 14.7620 | 116000 | 1.3013 |
| 14.8257 | 116500 | 1.298 |
| 14.8893 | 117000 | 1.2972 |
| 14.9529 | 117500 | 1.277 |
| 15.0165 | 118000 | 1.2718 |
| 15.0802 | 118500 | 1.1697 |
| 15.1438 | 119000 | 1.1819 |
| 15.2074 | 119500 | 1.1916 |
| 15.2711 | 120000 | 1.1829 |
| 15.3347 | 120500 | 1.1632 |
| 15.3983 | 121000 | 1.1809 |
| 15.4619 | 121500 | 1.1913 |
| 15.5256 | 122000 | 1.1916 |
| 15.5892 | 122500 | 1.1969 |
| 15.6528 | 123000 | 1.1929 |
| 15.7165 | 123500 | 1.2086 |
| 15.7801 | 124000 | 1.1864 |
| 15.8437 | 124500 | 1.2068 |
| 15.9074 | 125000 | 1.2253 |
| 15.9710 | 125500 | 1.1963 |
| 16.0346 | 126000 | 1.1585 |
| 16.0982 | 126500 | 1.0834 |
| 16.1619 | 127000 | 1.0937 |
| 16.2255 | 127500 | 1.0995 |
| 16.2891 | 128000 | 1.0787 |
| 16.3528 | 128500 | 1.1217 |
| 16.4164 | 129000 | 1.1185 |
| 16.4800 | 129500 | 1.1203 |
| 16.5436 | 130000 | 1.1201 |
| 16.6073 | 130500 | 1.125 |
| 16.6709 | 131000 | 1.1214 |
| 16.7345 | 131500 | 1.1228 |
| 16.7982 | 132000 | 1.1381 |
| 16.8618 | 132500 | 1.1414 |
| 16.9254 | 133000 | 1.123 |
| 16.9891 | 133500 | 1.1003 |
| 17.0527 | 134000 | 1.0447 |
| 17.1163 | 134500 | 1.036 |
| 17.1799 | 135000 | 1.0264 |
| 17.2436 | 135500 | 1.0375 |
| 17.3072 | 136000 | 1.0509 |
| 17.3708 | 136500 | 1.0452 |
| 17.4345 | 137000 | 1.0519 |
| 17.4981 | 137500 | 1.0498 |
| 17.5617 | 138000 | 1.0514 |
| 17.6253 | 138500 | 1.054 |
| 17.6890 | 139000 | 1.0457 |
| 17.7526 | 139500 | 1.0582 |
| 17.8162 | 140000 | 1.0566 |
| 17.8799 | 140500 | 1.0644 |
| 17.9435 | 141000 | 1.0579 |
| 18.0071 | 141500 | 1.0647 |
| 18.0708 | 142000 | 0.9704 |
| 18.1344 | 142500 | 0.9787 |
| 18.1980 | 143000 | 0.9875 |
| 18.2616 | 143500 | 0.987 |
| 18.3253 | 144000 | 0.9834 |
| 18.3889 | 144500 | 0.999 |
| 18.4525 | 145000 | 0.9872 |
| 18.5162 | 145500 | 0.9851 |
| 18.5798 | 146000 | 0.9986 |
| 18.6434 | 146500 | 0.9853 |
| 18.7071 | 147000 | 0.9973 |
| 18.7707 | 147500 | 0.988 |
| 18.8343 | 148000 | 0.999 |
| 18.8979 | 148500 | 0.9899 |
| 18.9616 | 149000 | 1.0053 |
| 19.0252 | 149500 | 0.9802 |
| 19.0888 | 150000 | 0.9301 |
| 19.1525 | 150500 | 0.9295 |
| 19.2161 | 151000 | 0.9334 |
| 19.2797 | 151500 | 0.9503 |
| 19.3433 | 152000 | 0.9161 |
| 19.4070 | 152500 | 0.9433 |
| 19.4706 | 153000 | 0.9376 |
| 19.5342 | 153500 | 0.9274 |
| 19.5979 | 154000 | 0.9414 |
| 19.6615 | 154500 | 0.94 |
| 19.7251 | 155000 | 0.9344 |
| 19.7888 | 155500 | 0.9464 |
| 19.8524 | 156000 | 0.9583 |
| 19.9160 | 156500 | 0.953 |
| 19.9796 | 157000 | 0.9481 |
| 20.0433 | 157500 | 0.8982 |
| 20.1069 | 158000 | 0.8974 |
| 20.1705 | 158500 | 0.9022 |
| 20.2342 | 159000 | 0.8923 |
| 20.2978 | 159500 | 0.8935 |
| 20.3614 | 160000 | 0.8917 |
| 20.4250 | 160500 | 0.9021 |
| 20.4887 | 161000 | 0.8978 |
| 20.5523 | 161500 | 0.9078 |
| 20.6159 | 162000 | 0.903 |
| 20.6796 | 162500 | 0.8989 |
| 20.7432 | 163000 | 0.9023 |
| 20.8068 | 163500 | 0.8918 |
| 20.8705 | 164000 | 0.8968 |
| 20.9341 | 164500 | 0.8977 |
| 20.9977 | 165000 | 0.9035 |
| 21.0613 | 165500 | 0.8347 |
| 21.1250 | 166000 | 0.8415 |
| 21.1886 | 166500 | 0.8472 |
| 21.2522 | 167000 | 0.8663 |
| 21.3159 | 167500 | 0.8633 |
| 21.3795 | 168000 | 0.8569 |
| 21.4431 | 168500 | 0.8529 |
| 21.5067 | 169000 | 0.8485 |
| 21.5704 | 169500 | 0.8759 |
| 21.6340 | 170000 | 0.8667 |
| 21.6976 | 170500 | 0.8615 |
| 21.7613 | 171000 | 0.8623 |
| 21.8249 | 171500 | 0.8613 |
| 21.8885 | 172000 | 0.8515 |
| 21.9522 | 172500 | 0.8615 |
| 22.0158 | 173000 | 0.8457 |
| 22.0794 | 173500 | 0.8106 |
| 22.1430 | 174000 | 0.8109 |
| 22.2067 | 174500 | 0.8108 |
| 22.2703 | 175000 | 0.8197 |
| 22.3339 | 175500 | 0.8165 |
| 22.3976 | 176000 | 0.8289 |
| 22.4612 | 176500 | 0.8288 |
| 22.5248 | 177000 | 0.8145 |
| 22.5884 | 177500 | 0.8249 |
| 22.6521 | 178000 | 0.8218 |
| 22.7157 | 178500 | 0.8284 |
| 22.7793 | 179000 | 0.833 |
| 22.8430 | 179500 | 0.8176 |
| 22.9066 | 180000 | 0.8431 |
| 22.9702 | 180500 | 0.8234 |
| 23.0339 | 181000 | 0.7998 |
| 23.0975 | 181500 | 0.7821 |
| 23.1611 | 182000 | 0.7914 |
| 23.2247 | 182500 | 0.7851 |
| 23.2884 | 183000 | 0.7797 |
| 23.3520 | 183500 | 0.7931 |
| 23.4156 | 184000 | 0.7912 |
| 23.4793 | 184500 | 0.7876 |
| 23.5429 | 185000 | 0.7954 |
| 23.6065 | 185500 | 0.7946 |
| 23.6701 | 186000 | 0.7782 |
| 23.7338 | 186500 | 0.7952 |
| 23.7974 | 187000 | 0.8015 |
| 23.8610 | 187500 | 0.7977 |
| 23.9247 | 188000 | 0.7875 |
| 23.9883 | 188500 | 0.7935 |
| 24.0519 | 189000 | 0.7617 |
| 24.1156 | 189500 | 0.7625 |
| 24.1792 | 190000 | 0.7514 |
| 24.2428 | 190500 | 0.7662 |
| 24.3064 | 191000 | 0.7692 |
| 24.3701 | 191500 | 0.7733 |
| 24.4337 | 192000 | 0.7561 |
| 24.4973 | 192500 | 0.7577 |
| 24.5610 | 193000 | 0.7687 |
| 24.6246 | 193500 | 0.7647 |
| 24.6882 | 194000 | 0.7717 |
| 24.7518 | 194500 | 0.761 |
| 24.8155 | 195000 | 0.7661 |
| 24.8791 | 195500 | 0.7446 |
| 24.9427 | 196000 | 0.7659 |
| 25.0064 | 196500 | 0.7559 |
| 25.0700 | 197000 | 0.7183 |
| 25.1336 | 197500 | 0.7399 |
| 25.1973 | 198000 | 0.7308 |
| 25.2609 | 198500 | 0.733 |
| 25.3245 | 199000 | 0.746 |
| 25.3881 | 199500 | 0.7274 |
| 25.4518 | 200000 | 0.7358 |
| 25.5154 | 200500 | 0.7468 |
| 25.5790 | 201000 | 0.734 |
| 25.6427 | 201500 | 0.7493 |
| 25.7063 | 202000 | 0.7263 |
| 25.7699 | 202500 | 0.7355 |
| 25.8335 | 203000 | 0.745 |
| 25.8972 | 203500 | 0.7301 |
| 25.9608 | 204000 | 0.7457 |
| 26.0244 | 204500 | 0.7072 |
| 26.0881 | 205000 | 0.7212 |
| 26.1517 | 205500 | 0.7186 |
| 26.2153 | 206000 | 0.7225 |
| 26.2790 | 206500 | 0.7065 |
| 26.3426 | 207000 | 0.7153 |
| 26.4062 | 207500 | 0.72 |
| 26.4698 | 208000 | 0.7074 |
| 26.5335 | 208500 | 0.7117 |
| 26.5971 | 209000 | 0.7206 |
| 26.6607 | 209500 | 0.7132 |
| 26.7244 | 210000 | 0.7199 |
| 26.7880 | 210500 | 0.7102 |
| 26.8516 | 211000 | 0.7155 |
| 26.9152 | 211500 | 0.7057 |
| 26.9789 | 212000 | 0.7191 |
| 27.0425 | 212500 | 0.6942 |
| 27.1061 | 213000 | 0.6924 |
| 27.1698 | 213500 | 0.7025 |
| 27.2334 | 214000 | 0.6911 |
| 27.2970 | 214500 | 0.6955 |
| 27.3607 | 215000 | 0.6875 |
| 27.4243 | 215500 | 0.698 |
| 27.4879 | 216000 | 0.7054 |
| 27.5515 | 216500 | 0.6968 |
| 27.6152 | 217000 | 0.7044 |
| 27.6788 | 217500 | 0.6946 |
| 27.7424 | 218000 | 0.6865 |
| 27.8061 | 218500 | 0.6974 |
| 27.8697 | 219000 | 0.698 |
| 27.9333 | 219500 | 0.6943 |
| 27.9969 | 220000 | 0.6985 |
| 28.0606 | 220500 | 0.6785 |
| 28.1242 | 221000 | 0.6842 |
| 28.1878 | 221500 | 0.6832 |
| 28.2515 | 222000 | 0.6863 |
| 28.3151 | 222500 | 0.6806 |
| 28.3787 | 223000 | 0.6897 |
| 28.4424 | 223500 | 0.6975 |
| 28.5060 | 224000 | 0.6802 |
| 28.5696 | 224500 | 0.6836 |
| 28.6332 | 225000 | 0.6849 |
| 28.6969 | 225500 | 0.6781 |
| 28.7605 | 226000 | 0.6761 |
| 28.8241 | 226500 | 0.6762 |
| 28.8878 | 227000 | 0.6781 |
| 28.9514 | 227500 | 0.682 |
| 29.0150 | 228000 | 0.6742 |
| 29.0786 | 228500 | 0.6595 |
| 29.1423 | 229000 | 0.683 |
| 29.2059 | 229500 | 0.6721 |
| 29.2695 | 230000 | 0.669 |
| 29.3332 | 230500 | 0.683 |
| 29.3968 | 231000 | 0.6652 |
| 29.4604 | 231500 | 0.671 |
| 29.5241 | 232000 | 0.6662 |
| 29.5877 | 232500 | 0.6665 |
| 29.6513 | 233000 | 0.6718 |
| 29.7149 | 233500 | 0.6657 |
| 29.7786 | 234000 | 0.6677 |
| 29.8422 | 234500 | 0.6732 |
| 29.9058 | 235000 | 0.6687 |
| 29.9695 | 235500 | 0.6732 |
Framework Versions
- Python: 3.11.5
- Sentence Transformers: 3.4.0
- Transformers: 4.48.0
- PyTorch: 2.5.1+cu124
- Accelerate: 1.2.1
- Datasets: 2.21.0
- Tokenizers: 0.21.0
Citation
BibTeX
Sentence Transformers
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
MarginMSELoss
@misc{hofstätter2021improving,
title={Improving Efficient Neural Ranking Models with Cross-Architecture Knowledge Distillation},
author={Sebastian Hofstätter and Sophia Althammer and Michael Schröder and Mete Sertkan and Allan Hanbury},
year={2021},
eprint={2010.02666},
archivePrefix={arXiv},
primaryClass={cs.IR}
}