tao-8k / README.md
Amu's picture
Update README.md
bc783a1
---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
model-index:
- name: tao-8k-origin
results:
- task:
type: STS
dataset:
type: C-MTEB/AFQMC
name: MTEB AFQMC
config: default
split: validation
revision: None
metrics:
- type: cos_sim_pearson
value: 47.33644889578121
- type: cos_sim_spearman
value: 49.93968642502866
- type: euclidean_pearson
value: 48.12029792973887
- type: euclidean_spearman
value: 49.939666315145494
- type: manhattan_pearson
value: 48.07449594650583
- type: manhattan_spearman
value: 49.892461433911166
- task:
type: STS
dataset:
type: C-MTEB/ATEC
name: MTEB ATEC
config: default
split: test
revision: None
metrics:
- type: cos_sim_pearson
value: 50.976148098905746
- type: cos_sim_spearman
value: 53.11230114448237
- type: euclidean_pearson
value: 55.119977161851054
- type: euclidean_spearman
value: 53.11229776647941
- type: manhattan_pearson
value: 55.096968162828034
- type: manhattan_spearman
value: 53.107481302419465
- task:
type: Classification
dataset:
type: mteb/amazon_reviews_multi
name: MTEB AmazonReviewsClassification (zh)
config: zh
split: test
revision: 1399c76144fd37290681b995c656ef9b2e06e26d
metrics:
- type: accuracy
value: 40.804
- type: f1
value: 39.01066543513968
- task:
type: STS
dataset:
type: C-MTEB/BQ
name: MTEB BQ
config: default
split: test
revision: None
metrics:
- type: cos_sim_pearson
value: 62.843816050026824
- type: cos_sim_spearman
value: 65.54142642656706
- type: euclidean_pearson
value: 64.08809634876388
- type: euclidean_spearman
value: 65.54142642558392
- type: manhattan_pearson
value: 64.09391522108272
- type: manhattan_spearman
value: 65.55445491162718
- task:
type: Clustering
dataset:
type: C-MTEB/CLSClusteringP2P
name: MTEB CLSClusteringP2P
config: default
split: test
revision: None
metrics:
- type: v_measure
value: 40.028061591547804
- task:
type: Clustering
dataset:
type: C-MTEB/CLSClusteringS2S
name: MTEB CLSClusteringS2S
config: default
split: test
revision: None
metrics:
- type: v_measure
value: 38.1897102944254
- task:
type: Reranking
dataset:
type: C-MTEB/CMedQAv1-reranking
name: MTEB CMedQAv1
config: default
split: test
revision: None
metrics:
- type: map
value: 85.34294439514511
- type: mrr
value: 88.03849206349206
- task:
type: Reranking
dataset:
type: C-MTEB/CMedQAv2-reranking
name: MTEB CMedQAv2
config: default
split: test
revision: None
metrics:
- type: map
value: 85.81294364673899
- type: mrr
value: 88.52146825396825
- task:
type: Retrieval
dataset:
type: C-MTEB/CmedqaRetrieval
name: MTEB CmedqaRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 23.982
- type: map_at_10
value: 36.21
- type: map_at_100
value: 38.072
- type: map_at_1000
value: 38.194
- type: map_at_3
value: 32.239000000000004
- type: map_at_5
value: 34.377
- type: mrr_at_1
value: 36.858999999999995
- type: mrr_at_10
value: 45.084999999999994
- type: mrr_at_100
value: 46.104
- type: mrr_at_1000
value: 46.154
- type: mrr_at_3
value: 42.623
- type: mrr_at_5
value: 43.995
- type: ndcg_at_1
value: 36.858999999999995
- type: ndcg_at_10
value: 42.735
- type: ndcg_at_100
value: 50.181
- type: ndcg_at_1000
value: 52.309000000000005
- type: ndcg_at_3
value: 37.728
- type: ndcg_at_5
value: 39.664
- type: precision_at_1
value: 36.858999999999995
- type: precision_at_10
value: 9.615
- type: precision_at_100
value: 1.564
- type: precision_at_1000
value: 0.183
- type: precision_at_3
value: 21.514
- type: precision_at_5
value: 15.568999999999999
- type: recall_at_1
value: 23.982
- type: recall_at_10
value: 53.04600000000001
- type: recall_at_100
value: 84.113
- type: recall_at_1000
value: 98.37
- type: recall_at_3
value: 37.824999999999996
- type: recall_at_5
value: 44.023
- task:
type: PairClassification
dataset:
type: C-MTEB/CMNLI
name: MTEB Cmnli
config: default
split: validation
revision: None
metrics:
- type: cos_sim_accuracy
value: 77.71497294046902
- type: cos_sim_ap
value: 86.84526989595028
- type: cos_sim_f1
value: 79.31987247608926
- type: cos_sim_precision
value: 72.70601987142022
- type: cos_sim_recall
value: 87.2574234276362
- type: dot_accuracy
value: 77.71497294046902
- type: dot_ap
value: 86.83880734247957
- type: dot_f1
value: 79.31987247608926
- type: dot_precision
value: 72.70601987142022
- type: dot_recall
value: 87.2574234276362
- type: euclidean_accuracy
value: 77.71497294046902
- type: euclidean_ap
value: 86.84526869685902
- type: euclidean_f1
value: 79.31987247608926
- type: euclidean_precision
value: 72.70601987142022
- type: euclidean_recall
value: 87.2574234276362
- type: manhattan_accuracy
value: 77.8111846061335
- type: manhattan_ap
value: 86.81142881585656
- type: manhattan_f1
value: 79.4201671780764
- type: manhattan_precision
value: 72.53575570158485
- type: manhattan_recall
value: 87.74842179097499
- type: max_accuracy
value: 77.8111846061335
- type: max_ap
value: 86.84526989595028
- type: max_f1
value: 79.4201671780764
- task:
type: Retrieval
dataset:
type: C-MTEB/CovidRetrieval
name: MTEB CovidRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 70.706
- type: map_at_10
value: 78.619
- type: map_at_100
value: 78.915
- type: map_at_1000
value: 78.918
- type: map_at_3
value: 76.967
- type: map_at_5
value: 77.922
- type: mrr_at_1
value: 70.917
- type: mrr_at_10
value: 78.64
- type: mrr_at_100
value: 78.935
- type: mrr_at_1000
value: 78.938
- type: mrr_at_3
value: 77.081
- type: mrr_at_5
value: 77.972
- type: ndcg_at_1
value: 70.917
- type: ndcg_at_10
value: 82.186
- type: ndcg_at_100
value: 83.487
- type: ndcg_at_1000
value: 83.589
- type: ndcg_at_3
value: 78.874
- type: ndcg_at_5
value: 80.548
- type: precision_at_1
value: 70.917
- type: precision_at_10
value: 9.431000000000001
- type: precision_at_100
value: 1.001
- type: precision_at_1000
value: 0.101
- type: precision_at_3
value: 28.275
- type: precision_at_5
value: 17.829
- type: recall_at_1
value: 70.706
- type: recall_at_10
value: 93.256
- type: recall_at_100
value: 99.05199999999999
- type: recall_at_1000
value: 99.895
- type: recall_at_3
value: 84.247
- type: recall_at_5
value: 88.251
- task:
type: Retrieval
dataset:
type: C-MTEB/DuRetrieval
name: MTEB DuRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 25.989
- type: map_at_10
value: 80.882
- type: map_at_100
value: 83.63199999999999
- type: map_at_1000
value: 83.663
- type: map_at_3
value: 55.772
- type: map_at_5
value: 70.598
- type: mrr_at_1
value: 90.14999999999999
- type: mrr_at_10
value: 93.30000000000001
- type: mrr_at_100
value: 93.363
- type: mrr_at_1000
value: 93.366
- type: mrr_at_3
value: 93.083
- type: mrr_at_5
value: 93.206
- type: ndcg_at_1
value: 90.14999999999999
- type: ndcg_at_10
value: 88.016
- type: ndcg_at_100
value: 90.52900000000001
- type: ndcg_at_1000
value: 90.84400000000001
- type: ndcg_at_3
value: 86.529
- type: ndcg_at_5
value: 85.65899999999999
- type: precision_at_1
value: 90.14999999999999
- type: precision_at_10
value: 42.295
- type: precision_at_100
value: 4.826
- type: precision_at_1000
value: 0.48900000000000005
- type: precision_at_3
value: 77.717
- type: precision_at_5
value: 65.81
- type: recall_at_1
value: 25.989
- type: recall_at_10
value: 89.446
- type: recall_at_100
value: 97.832
- type: recall_at_1000
value: 99.568
- type: recall_at_3
value: 58.223
- type: recall_at_5
value: 75.411
- task:
type: Retrieval
dataset:
type: C-MTEB/EcomRetrieval
name: MTEB EcomRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 49.6
- type: map_at_10
value: 59.512
- type: map_at_100
value: 60.059
- type: map_at_1000
value: 60.077999999999996
- type: map_at_3
value: 56.882999999999996
- type: map_at_5
value: 58.298
- type: mrr_at_1
value: 49.6
- type: mrr_at_10
value: 59.512
- type: mrr_at_100
value: 60.059
- type: mrr_at_1000
value: 60.077999999999996
- type: mrr_at_3
value: 56.882999999999996
- type: mrr_at_5
value: 58.298
- type: ndcg_at_1
value: 49.6
- type: ndcg_at_10
value: 64.71000000000001
- type: ndcg_at_100
value: 67.238
- type: ndcg_at_1000
value: 67.74
- type: ndcg_at_3
value: 59.275
- type: ndcg_at_5
value: 61.805
- type: precision_at_1
value: 49.6
- type: precision_at_10
value: 8.12
- type: precision_at_100
value: 0.927
- type: precision_at_1000
value: 0.097
- type: precision_at_3
value: 22.067
- type: precision_at_5
value: 14.46
- type: recall_at_1
value: 49.6
- type: recall_at_10
value: 81.2
- type: recall_at_100
value: 92.7
- type: recall_at_1000
value: 96.6
- type: recall_at_3
value: 66.2
- type: recall_at_5
value: 72.3
- task:
type: Classification
dataset:
type: C-MTEB/IFlyTek-classification
name: MTEB IFlyTek
config: default
split: validation
revision: None
metrics:
- type: accuracy
value: 47.98768757214313
- type: f1
value: 35.24243089488371
- task:
type: Classification
dataset:
type: C-MTEB/JDReview-classification
name: MTEB JDReview
config: default
split: test
revision: None
metrics:
- type: accuracy
value: 87.01688555347093
- type: ap
value: 56.39167630414159
- type: f1
value: 81.91756262306008
- task:
type: STS
dataset:
type: C-MTEB/LCQMC
name: MTEB LCQMC
config: default
split: test
revision: None
metrics:
- type: cos_sim_pearson
value: 71.17874301231225
- type: cos_sim_spearman
value: 77.47936067899236
- type: euclidean_pearson
value: 76.3241109984839
- type: euclidean_spearman
value: 77.47936511149533
- type: manhattan_pearson
value: 76.3334642249198
- type: manhattan_spearman
value: 77.48889610190774
- task:
type: Reranking
dataset:
type: C-MTEB/Mmarco-reranking
name: MTEB MMarcoReranking
config: default
split: dev
revision: None
metrics:
- type: map
value: 27.96872431410137
- type: mrr
value: 26.92023809523809
- task:
type: Retrieval
dataset:
type: C-MTEB/MMarcoRetrieval
name: MTEB MMarcoRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 66.83099999999999
- type: map_at_10
value: 75.945
- type: map_at_100
value: 76.259
- type: map_at_1000
value: 76.27000000000001
- type: map_at_3
value: 74.22999999999999
- type: map_at_5
value: 75.318
- type: mrr_at_1
value: 69.069
- type: mrr_at_10
value: 76.491
- type: mrr_at_100
value: 76.764
- type: mrr_at_1000
value: 76.775
- type: mrr_at_3
value: 75.01
- type: mrr_at_5
value: 75.934
- type: ndcg_at_1
value: 69.069
- type: ndcg_at_10
value: 79.557
- type: ndcg_at_100
value: 80.946
- type: ndcg_at_1000
value: 81.23700000000001
- type: ndcg_at_3
value: 76.31099999999999
- type: ndcg_at_5
value: 78.121
- type: precision_at_1
value: 69.069
- type: precision_at_10
value: 9.58
- type: precision_at_100
value: 1.027
- type: precision_at_1000
value: 0.105
- type: precision_at_3
value: 28.73
- type: precision_at_5
value: 18.201
- type: recall_at_1
value: 66.83099999999999
- type: recall_at_10
value: 90.118
- type: recall_at_100
value: 96.377
- type: recall_at_1000
value: 98.656
- type: recall_at_3
value: 81.516
- type: recall_at_5
value: 85.798
- task:
type: Classification
dataset:
type: mteb/amazon_massive_intent
name: MTEB MassiveIntentClassification (zh-CN)
config: zh-CN
split: test
revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7
metrics:
- type: accuracy
value: 68.2649630127774
- type: f1
value: 65.96868218344183
- task:
type: Classification
dataset:
type: mteb/amazon_massive_scenario
name: MTEB MassiveScenarioClassification (zh-CN)
config: zh-CN
split: test
revision: 7d571f92784cd94a019292a1f45445077d0ef634
metrics:
- type: accuracy
value: 73.13382649630127
- type: f1
value: 72.69980239148315
- task:
type: Retrieval
dataset:
type: C-MTEB/MedicalRetrieval
name: MTEB MedicalRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 51.2
- type: map_at_10
value: 57.715
- type: map_at_100
value: 58.233999999999995
- type: map_at_1000
value: 58.289
- type: map_at_3
value: 56.483000000000004
- type: map_at_5
value: 57.193000000000005
- type: mrr_at_1
value: 51.2
- type: mrr_at_10
value: 57.714
- type: mrr_at_100
value: 58.233000000000004
- type: mrr_at_1000
value: 58.288
- type: mrr_at_3
value: 56.483000000000004
- type: mrr_at_5
value: 57.193000000000005
- type: ndcg_at_1
value: 51.2
- type: ndcg_at_10
value: 60.63499999999999
- type: ndcg_at_100
value: 63.458000000000006
- type: ndcg_at_1000
value: 64.992
- type: ndcg_at_3
value: 58.11300000000001
- type: ndcg_at_5
value: 59.391000000000005
- type: precision_at_1
value: 51.2
- type: precision_at_10
value: 6.97
- type: precision_at_100
value: 0.836
- type: precision_at_1000
value: 0.096
- type: precision_at_3
value: 20.933
- type: precision_at_5
value: 13.18
- type: recall_at_1
value: 51.2
- type: recall_at_10
value: 69.69999999999999
- type: recall_at_100
value: 83.6
- type: recall_at_1000
value: 95.8
- type: recall_at_3
value: 62.8
- type: recall_at_5
value: 65.9
- task:
type: Classification
dataset:
type: C-MTEB/MultilingualSentiment-classification
name: MTEB MultilingualSentiment
config: default
split: validation
revision: None
metrics:
- type: accuracy
value: 73.39
- type: f1
value: 72.85739851837214
- task:
type: PairClassification
dataset:
type: C-MTEB/OCNLI
name: MTEB Ocnli
config: default
split: validation
revision: None
metrics:
- type: cos_sim_accuracy
value: 73.36220898754738
- type: cos_sim_ap
value: 78.50045169678386
- type: cos_sim_f1
value: 75.3875968992248
- type: cos_sim_precision
value: 69.65085049239033
- type: cos_sim_recall
value: 82.15417106652588
- type: dot_accuracy
value: 73.36220898754738
- type: dot_ap
value: 78.50039148302838
- type: dot_f1
value: 75.3875968992248
- type: dot_precision
value: 69.65085049239033
- type: dot_recall
value: 82.15417106652588
- type: euclidean_accuracy
value: 73.36220898754738
- type: euclidean_ap
value: 78.50045169678386
- type: euclidean_f1
value: 75.3875968992248
- type: euclidean_precision
value: 69.65085049239033
- type: euclidean_recall
value: 82.15417106652588
- type: manhattan_accuracy
value: 73.09149972929075
- type: manhattan_ap
value: 78.40911589236852
- type: manhattan_f1
value: 75.3623188405797
- type: manhattan_precision
value: 69.45681211041853
- type: manhattan_recall
value: 82.36536430834214
- type: max_accuracy
value: 73.36220898754738
- type: max_ap
value: 78.50045169678386
- type: max_f1
value: 75.3875968992248
- task:
type: Classification
dataset:
type: C-MTEB/OnlineShopping-classification
name: MTEB OnlineShopping
config: default
split: test
revision: None
metrics:
- type: accuracy
value: 91.81000000000002
- type: ap
value: 89.35809579688139
- type: f1
value: 91.79220350456818
- task:
type: STS
dataset:
type: C-MTEB/PAWSX
name: MTEB PAWSX
config: default
split: test
revision: None
metrics:
- type: cos_sim_pearson
value: 30.06960208048424
- type: cos_sim_spearman
value: 36.21568893707218
- type: euclidean_pearson
value: 36.3789158810154
- type: euclidean_spearman
value: 36.21568740241203
- type: manhattan_pearson
value: 36.318190228955935
- type: manhattan_spearman
value: 36.16813420759451
- task:
type: STS
dataset:
type: C-MTEB/QBQTC
name: MTEB QBQTC
config: default
split: test
revision: None
metrics:
- type: cos_sim_pearson
value: 36.779942621488736
- type: cos_sim_spearman
value: 38.73716529566492
- type: euclidean_pearson
value: 37.134107612179605
- type: euclidean_spearman
value: 38.737099842399545
- type: manhattan_pearson
value: 37.17579625045808
- type: manhattan_spearman
value: 38.746051563332315
- task:
type: STS
dataset:
type: mteb/sts22-crosslingual-sts
name: MTEB STS22 (zh)
config: zh
split: test
revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80
metrics:
- type: cos_sim_pearson
value: 65.97416499132073
- type: cos_sim_spearman
value: 68.87894646940939
- type: euclidean_pearson
value: 67.2366929400408
- type: euclidean_spearman
value: 68.87894646940939
- type: manhattan_pearson
value: 67.30590304353478
- type: manhattan_spearman
value: 68.90546655032796
- task:
type: STS
dataset:
type: C-MTEB/STSB
name: MTEB STSB
config: default
split: test
revision: None
metrics:
- type: cos_sim_pearson
value: 78.99420906581649
- type: cos_sim_spearman
value: 79.36553449000968
- type: euclidean_pearson
value: 78.77734144763518
- type: euclidean_spearman
value: 79.36545230850567
- type: manhattan_pearson
value: 78.82512507141092
- type: manhattan_spearman
value: 79.43977311125059
- task:
type: Reranking
dataset:
type: C-MTEB/T2Reranking
name: MTEB T2Reranking
config: default
split: dev
revision: None
metrics:
- type: map
value: 66.38018284846501
- type: mrr
value: 76.11180965277104
- task:
type: Retrieval
dataset:
type: C-MTEB/T2Retrieval
name: MTEB T2Retrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 27.423
- type: map_at_10
value: 77.206
- type: map_at_100
value: 80.83500000000001
- type: map_at_1000
value: 80.9
- type: map_at_3
value: 54.190000000000005
- type: map_at_5
value: 66.662
- type: mrr_at_1
value: 90.049
- type: mrr_at_10
value: 92.48100000000001
- type: mrr_at_100
value: 92.567
- type: mrr_at_1000
value: 92.571
- type: mrr_at_3
value: 92.07
- type: mrr_at_5
value: 92.32900000000001
- type: ndcg_at_1
value: 90.049
- type: ndcg_at_10
value: 84.69
- type: ndcg_at_100
value: 88.254
- type: ndcg_at_1000
value: 88.89399999999999
- type: ndcg_at_3
value: 86.091
- type: ndcg_at_5
value: 84.685
- type: precision_at_1
value: 90.049
- type: precision_at_10
value: 42.141
- type: precision_at_100
value: 5.016
- type: precision_at_1000
value: 0.516
- type: precision_at_3
value: 75.352
- type: precision_at_5
value: 63.176
- type: recall_at_1
value: 27.423
- type: recall_at_10
value: 83.595
- type: recall_at_100
value: 95.21
- type: recall_at_1000
value: 98.503
- type: recall_at_3
value: 55.84400000000001
- type: recall_at_5
value: 69.987
- task:
type: Classification
dataset:
type: C-MTEB/TNews-classification
name: MTEB TNews
config: default
split: validation
revision: None
metrics:
- type: accuracy
value: 51.927
- type: f1
value: 50.16838216110367
- task:
type: Clustering
dataset:
type: C-MTEB/ThuNewsClusteringP2P
name: MTEB ThuNewsClusteringP2P
config: default
split: test
revision: None
metrics:
- type: v_measure
value: 60.85131720842154
- task:
type: Clustering
dataset:
type: C-MTEB/ThuNewsClusteringS2S
name: MTEB ThuNewsClusteringS2S
config: default
split: test
revision: None
metrics:
- type: v_measure
value: 57.0921610946628
- task:
type: Retrieval
dataset:
type: C-MTEB/VideoRetrieval
name: MTEB VideoRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 56.99999999999999
- type: map_at_10
value: 67.611
- type: map_at_100
value: 68.095
- type: map_at_1000
value: 68.10300000000001
- type: map_at_3
value: 65.75
- type: map_at_5
value: 66.93
- type: mrr_at_1
value: 56.89999999999999
- type: mrr_at_10
value: 67.561
- type: mrr_at_100
value: 68.045
- type: mrr_at_1000
value: 68.053
- type: mrr_at_3
value: 65.7
- type: mrr_at_5
value: 66.88
- type: ndcg_at_1
value: 56.99999999999999
- type: ndcg_at_10
value: 72.25200000000001
- type: ndcg_at_100
value: 74.542
- type: ndcg_at_1000
value: 74.725
- type: ndcg_at_3
value: 68.47
- type: ndcg_at_5
value: 70.583
- type: precision_at_1
value: 56.99999999999999
- type: precision_at_10
value: 8.66
- type: precision_at_100
value: 0.972
- type: precision_at_1000
value: 0.099
- type: precision_at_3
value: 25.433
- type: precision_at_5
value: 16.28
- type: recall_at_1
value: 56.99999999999999
- type: recall_at_10
value: 86.6
- type: recall_at_100
value: 97.2
- type: recall_at_1000
value: 98.6
- type: recall_at_3
value: 76.3
- type: recall_at_5
value: 81.39999999999999
- task:
type: Classification
dataset:
type: C-MTEB/waimai-classification
name: MTEB Waimai
config: default
split: test
revision: None
metrics:
- type: accuracy
value: 87.10000000000001
- type: ap
value: 70.81766065881429
- type: f1
value: 85.5323306120456
license: apache-2.0
language:
- zh
---
A try for emebdding model:
The method is the same as the stella-v2, I just extend the length of the context on tao.(I found if you want to use the fully-8k context, you maybe need to convert the model to float32).
Now I'm working on the tao-v2, It will have a different sturcture.
I will release tao-v2 as fast as I can.
Thank you to the open source community.