--- model-index: - name: Yuan-embedding-1.0 results: - dataset: config: default name: MTEB AFQMC (default) revision: None split: validation type: C-MTEB/AFQMC metrics: - type: cosine_pearson value: 56.398777687800596 - type: cosine_spearman value: 60.2976392017466 - type: manhattan_pearson value: 58.34432755369896 - type: manhattan_spearman value: 59.633715024557176 - type: euclidean_pearson value: 58.33199470250656 - type: euclidean_spearman value: 59.633393360323595 - type: main_score value: 60.2976392017466 task: type: STS - dataset: config: default name: MTEB ATEC (default) revision: None split: test type: C-MTEB/ATEC metrics: - type: cosine_pearson value: 56.418711941754694 - type: cosine_spearman value: 58.49782527525838 - type: manhattan_pearson value: 62.05335398720773 - type: manhattan_spearman value: 58.18176592298454 - type: euclidean_pearson value: 62.06479799788818 - type: euclidean_spearman value: 58.18182671971488 - type: main_score value: 58.49782527525838 task: type: STS - dataset: config: zh name: MTEB AmazonReviewsClassification (zh) revision: 1399c76144fd37290681b995c656ef9b2e06e26d split: test type: mteb/amazon_reviews_multi metrics: - type: accuracy value: 46.656000000000006 - type: accuracy_stderr value: 1.1704631561907444 - type: f1 value: 45.75911645865614 - type: f1_stderr value: 1.323301406018355 - type: main_score value: 46.656000000000006 task: type: Classification - dataset: config: zh name: MTEB AmazonReviewsClassification (zh) revision: 1399c76144fd37290681b995c656ef9b2e06e26d split: validation type: mteb/amazon_reviews_multi metrics: - type: accuracy value: 45.84599999999999 - type: accuracy_stderr value: 1.0539468677310073 - type: f1 value: 45.03273670979488 - type: f1_stderr value: 1.00417269917164 - type: main_score value: 45.84599999999999 task: type: Classification - dataset: config: default name: MTEB BQ (default) revision: None split: test type: C-MTEB/BQ metrics: - type: cosine_pearson value: 71.33099160181597 - type: cosine_spearman value: 73.06963287952199 - type: manhattan_pearson value: 70.65314181752566 - type: manhattan_spearman value: 72.34604440078336 - type: euclidean_pearson value: 70.67624292501411 - type: euclidean_spearman value: 72.3597691712343 - type: main_score value: 73.06963287952199 task: type: STS - dataset: config: default name: MTEB CLSClusteringP2P (default) revision: None split: test type: C-MTEB/CLSClusteringP2P metrics: - type: v_measure value: 53.79921861868626 - type: v_measure_std value: 2.073016548125077 - type: main_score value: 53.79921861868626 task: type: Clustering - dataset: config: default name: MTEB CLSClusteringS2S (default) revision: None split: test type: C-MTEB/CLSClusteringS2S metrics: - type: v_measure value: 46.22496957569903 - type: v_measure_std value: 1.4660184854965337 - type: main_score value: 46.22496957569903 task: type: Clustering - dataset: config: default name: MTEB CMedQAv1-reranking (default) revision: None split: test type: C-MTEB/CMedQAv1-reranking metrics: - type: map value: 90.00883554654739 - type: mrr value: 92.02547619047618 - type: main_score value: 90.00883554654739 task: type: Reranking - dataset: config: default name: MTEB CMedQAv2-reranking (default) revision: None split: test type: C-MTEB/CMedQAv2-reranking metrics: - type: map value: 92.47561424216632 - type: mrr value: 94.60039682539681 - type: main_score value: 92.47561424216632 task: type: Reranking - dataset: config: default name: MTEB CmedqaRetrieval (default) revision: None split: dev type: C-MTEB/CmedqaRetrieval metrics: - type: map_at_1 value: 29.935000000000002 - type: map_at_10 value: 44.143 - type: map_at_100 value: 45.999 - type: map_at_1000 value: 46.084 - type: map_at_3 value: 39.445 - type: map_at_5 value: 42.218 - type: mrr_at_1 value: 44.711 - type: mrr_at_10 value: 53.88699999999999 - type: mrr_at_100 value: 54.813 - type: mrr_at_1000 value: 54.834 - type: mrr_at_3 value: 51.1 - type: mrr_at_5 value: 52.827 - type: ndcg_at_1 value: 44.711 - type: ndcg_at_10 value: 51.471999999999994 - type: ndcg_at_100 value: 58.362 - type: ndcg_at_1000 value: 59.607 - type: ndcg_at_3 value: 45.558 - type: ndcg_at_5 value: 48.345 - type: precision_at_1 value: 44.711 - type: precision_at_10 value: 11.1 - type: precision_at_100 value: 1.6650000000000003 - type: precision_at_1000 value: 0.184 - type: precision_at_3 value: 25.306 - type: precision_at_5 value: 18.404999999999998 - type: recall_at_1 value: 29.935000000000002 - type: recall_at_10 value: 63.366 - type: recall_at_100 value: 91.375 - type: recall_at_1000 value: 99.167 - type: recall_at_3 value: 45.888 - type: recall_at_5 value: 54.169 - type: main_score value: 51.471999999999994 task: type: Retrieval - dataset: config: default name: MTEB Cmnli (default) revision: None split: validation type: C-MTEB/CMNLI metrics: - type: cos_sim_accuracy value: 80.3968731208659 - type: cos_sim_accuracy_threshold value: 86.61384582519531 - type: cos_sim_ap value: 88.21894124132636 - type: cos_sim_f1 value: 81.67308750687947 - type: cos_sim_f1_threshold value: 86.04017496109009 - type: cos_sim_precision value: 77.1630615640599 - type: cos_sim_recall value: 86.7430441898527 - type: dot_accuracy value: 67.7931449188214 - type: dot_accuracy_threshold value: 92027.47802734375 - type: dot_ap value: 75.73048600318765 - type: dot_f1 value: 71.64554512914772 - type: dot_f1_threshold value: 83535.70556640625 - type: dot_precision value: 61.1056105610561 - type: dot_recall value: 86.57937806873977 - type: euclidean_accuracy value: 78.52074564040889 - type: euclidean_accuracy_threshold value: 1688.486671447754 - type: euclidean_ap value: 86.40643721988414 - type: euclidean_f1 value: 79.97822536744692 - type: euclidean_f1_threshold value: 1748.1914520263672 - type: euclidean_precision value: 74.83700081499592 - type: euclidean_recall value: 85.87795183539865 - type: manhattan_accuracy value: 78.59290438965725 - type: manhattan_accuracy_threshold value: 57066.162109375 - type: manhattan_ap value: 86.38300352696045 - type: manhattan_f1 value: 79.84587391630097 - type: manhattan_f1_threshold value: 59686.376953125 - type: manhattan_precision value: 73.62810896170548 - type: manhattan_recall value: 87.21066167874679 - type: max_accuracy value: 80.3968731208659 - type: max_ap value: 88.21894124132636 - type: max_f1 value: 81.67308750687947 task: type: PairClassification - dataset: config: default name: MTEB CovidRetrieval (default) revision: None split: dev type: C-MTEB/CovidRetrieval metrics: - type: map_at_1 value: 85.485 - type: map_at_10 value: 91.135 - type: map_at_100 value: 91.16199999999999 - type: map_at_1000 value: 91.16300000000001 - type: map_at_3 value: 90.499 - type: map_at_5 value: 90.91 - type: mrr_at_1 value: 85.88 - type: mrr_at_10 value: 91.133 - type: mrr_at_100 value: 91.16 - type: mrr_at_1000 value: 91.161 - type: mrr_at_3 value: 90.551 - type: mrr_at_5 value: 90.904 - type: ndcg_at_1 value: 85.88 - type: ndcg_at_10 value: 93.163 - type: ndcg_at_100 value: 93.282 - type: ndcg_at_1000 value: 93.309 - type: ndcg_at_3 value: 91.943 - type: ndcg_at_5 value: 92.637 - type: precision_at_1 value: 85.88 - type: precision_at_10 value: 10.032 - type: precision_at_100 value: 1.008 - type: precision_at_1000 value: 0.101 - type: precision_at_3 value: 32.315 - type: precision_at_5 value: 19.747 - type: recall_at_1 value: 85.485 - type: recall_at_10 value: 99.262 - type: recall_at_100 value: 99.789 - type: recall_at_1000 value: 100.0 - type: recall_at_3 value: 95.96900000000001 - type: recall_at_5 value: 97.682 - type: main_score value: 93.163 task: type: Retrieval - dataset: config: default name: MTEB DuRetrieval (default) revision: None split: dev type: C-MTEB/DuRetrieval metrics: - type: map_at_1 value: 27.29 - type: map_at_10 value: 82.832 - type: map_at_100 value: 85.482 - type: map_at_1000 value: 85.52 - type: map_at_3 value: 57.964000000000006 - type: map_at_5 value: 72.962 - type: mrr_at_1 value: 92.35 - type: mrr_at_10 value: 94.77499999999999 - type: mrr_at_100 value: 94.825 - type: mrr_at_1000 value: 94.827 - type: mrr_at_3 value: 94.50800000000001 - type: mrr_at_5 value: 94.688 - type: ndcg_at_1 value: 92.35 - type: ndcg_at_10 value: 89.432 - type: ndcg_at_100 value: 91.813 - type: ndcg_at_1000 value: 92.12 - type: ndcg_at_3 value: 88.804 - type: ndcg_at_5 value: 87.681 - type: precision_at_1 value: 92.35 - type: precision_at_10 value: 42.32 - type: precision_at_100 value: 4.812 - type: precision_at_1000 value: 0.48900000000000005 - type: precision_at_3 value: 79.367 - type: precision_at_5 value: 66.86999999999999 - type: recall_at_1 value: 27.29 - type: recall_at_10 value: 90.093 - type: recall_at_100 value: 97.916 - type: recall_at_1000 value: 99.40299999999999 - type: recall_at_3 value: 59.816 - type: recall_at_5 value: 76.889 - type: main_score value: 89.432 task: type: Retrieval - dataset: config: default name: MTEB EcomRetrieval (default) revision: None split: dev type: C-MTEB/EcomRetrieval metrics: - type: map_at_1 value: 55.2 - type: map_at_10 value: 65.767 - type: map_at_100 value: 66.208 - type: map_at_1000 value: 66.219 - type: map_at_3 value: 63.1 - type: map_at_5 value: 64.865 - type: mrr_at_1 value: 55.2 - type: mrr_at_10 value: 65.767 - type: mrr_at_100 value: 66.208 - type: mrr_at_1000 value: 66.219 - type: mrr_at_3 value: 63.1 - type: mrr_at_5 value: 64.865 - type: ndcg_at_1 value: 55.2 - type: ndcg_at_10 value: 70.875 - type: ndcg_at_100 value: 72.931 - type: ndcg_at_1000 value: 73.2 - type: ndcg_at_3 value: 65.526 - type: ndcg_at_5 value: 68.681 - type: precision_at_1 value: 55.2 - type: precision_at_10 value: 8.690000000000001 - type: precision_at_100 value: 0.963 - type: precision_at_1000 value: 0.098 - type: precision_at_3 value: 24.166999999999998 - type: precision_at_5 value: 16.02 - type: recall_at_1 value: 55.2 - type: recall_at_10 value: 86.9 - type: recall_at_100 value: 96.3 - type: recall_at_1000 value: 98.4 - type: recall_at_3 value: 72.5 - type: recall_at_5 value: 80.10000000000001 - type: main_score value: 70.875 task: type: Retrieval - dataset: config: default name: MTEB IFlyTek (default) revision: None split: validation type: C-MTEB/IFlyTek-classification metrics: - type: accuracy value: 46.95652173913043 - type: accuracy_stderr value: 0.8816372193041417 - type: f1 value: 38.870262239396496 - type: f1_stderr value: 1.1248427890133785 - type: main_score value: 46.95652173913043 task: type: Classification - dataset: config: default name: MTEB JDReview (default) revision: None split: test type: C-MTEB/JDReview-classification metrics: - type: accuracy value: 87.18574108818011 - type: accuracy_stderr value: 1.828763099528331 - type: ap value: 56.516251295719414 - type: ap_stderr value: 3.3789918068717895 - type: f1 value: 82.04209146803106 - type: f1_stderr value: 2.005027201503808 - type: main_score value: 87.18574108818011 task: type: Classification - dataset: config: default name: MTEB LCQMC (default) revision: None split: test type: C-MTEB/LCQMC metrics: - type: cosine_pearson value: 72.67112275922743 - type: cosine_spearman value: 78.44376213964316 - type: manhattan_pearson value: 77.51766838932976 - type: manhattan_spearman value: 78.02885255071602 - type: euclidean_pearson value: 77.5292348074114 - type: euclidean_spearman value: 78.04277103380235 - type: main_score value: 78.44376213964316 task: type: STS - dataset: config: default name: MTEB MMarcoReranking (default) revision: None split: dev type: C-MTEB/Mmarco-reranking metrics: - type: map value: 37.021133625346174 - type: mrr value: 35.81428571428572 - type: main_score value: 37.021133625346174 task: type: Reranking - dataset: config: default name: MTEB MMarcoRetrieval (default) revision: None split: dev type: C-MTEB/MMarcoRetrieval metrics: - type: map_at_1 value: 69.624 - type: map_at_10 value: 78.764 - type: map_at_100 value: 79.038 - type: map_at_1000 value: 79.042 - type: map_at_3 value: 76.846 - type: map_at_5 value: 78.106 - type: mrr_at_1 value: 71.905 - type: mrr_at_10 value: 79.268 - type: mrr_at_100 value: 79.508 - type: mrr_at_1000 value: 79.512 - type: mrr_at_3 value: 77.60000000000001 - type: mrr_at_5 value: 78.701 - type: ndcg_at_1 value: 71.905 - type: ndcg_at_10 value: 82.414 - type: ndcg_at_100 value: 83.59 - type: ndcg_at_1000 value: 83.708 - type: ndcg_at_3 value: 78.803 - type: ndcg_at_5 value: 80.94 - type: precision_at_1 value: 71.905 - type: precision_at_10 value: 9.901 - type: precision_at_100 value: 1.048 - type: precision_at_1000 value: 0.106 - type: precision_at_3 value: 29.479 - type: precision_at_5 value: 18.828 - type: recall_at_1 value: 69.624 - type: recall_at_10 value: 93.149 - type: recall_at_100 value: 98.367 - type: recall_at_1000 value: 99.29299999999999 - type: recall_at_3 value: 83.67599999999999 - type: recall_at_5 value: 88.752 - type: main_score value: 82.414 task: type: Retrieval - dataset: config: zh-CN name: MTEB MassiveIntentClassification (zh-CN) revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7 split: test type: mteb/amazon_massive_intent metrics: - type: accuracy value: 77.36045729657029 - type: accuracy_stderr value: 0.8944498935111289 - type: f1 value: 73.73485209304225 - type: f1_stderr value: 0.8615191738484445 - type: main_score value: 77.36045729657029 task: type: Classification - dataset: config: zh-CN name: MTEB MassiveIntentClassification (zh-CN) revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7 split: validation type: mteb/amazon_massive_intent metrics: - type: accuracy value: 78.16035415641909 - type: accuracy_stderr value: 0.7514724220154535 - type: f1 value: 75.32402452596266 - type: f1_stderr value: 0.5969737694527888 - type: main_score value: 78.16035415641909 task: type: Classification - dataset: config: zh-CN name: MTEB MassiveScenarioClassification (zh-CN) revision: 7d571f92784cd94a019292a1f45445077d0ef634 split: test type: mteb/amazon_massive_scenario metrics: - type: accuracy value: 83.31203765971755 - type: accuracy_stderr value: 1.1063564012537301 - type: f1 value: 82.81655735858999 - type: f1_stderr value: 0.9643568609098954 - type: main_score value: 83.31203765971755 task: type: Classification - dataset: config: zh-CN name: MTEB MassiveScenarioClassification (zh-CN) revision: 7d571f92784cd94a019292a1f45445077d0ef634 split: validation type: mteb/amazon_massive_scenario metrics: - type: accuracy value: 83.11362518445647 - type: accuracy_stderr value: 1.252141689154366 - type: f1 value: 82.56555569957769 - type: f1_stderr value: 0.858322314243248 - type: main_score value: 83.11362518445647 task: type: Classification - dataset: config: default name: MTEB MedicalRetrieval (default) revision: None split: dev type: C-MTEB/MedicalRetrieval metrics: - type: map_at_1 value: 63.1 - type: map_at_10 value: 70.816 - type: map_at_100 value: 71.368 - type: map_at_1000 value: 71.379 - type: map_at_3 value: 69.033 - type: map_at_5 value: 70.028 - type: mrr_at_1 value: 63.4 - type: mrr_at_10 value: 70.98400000000001 - type: mrr_at_100 value: 71.538 - type: mrr_at_1000 value: 71.548 - type: mrr_at_3 value: 69.19999999999999 - type: mrr_at_5 value: 70.195 - type: ndcg_at_1 value: 63.1 - type: ndcg_at_10 value: 74.665 - type: ndcg_at_100 value: 77.16199999999999 - type: ndcg_at_1000 value: 77.408 - type: ndcg_at_3 value: 70.952 - type: ndcg_at_5 value: 72.776 - type: precision_at_1 value: 63.1 - type: precision_at_10 value: 8.68 - type: precision_at_100 value: 0.9809999999999999 - type: precision_at_1000 value: 0.1 - type: precision_at_3 value: 25.5 - type: precision_at_5 value: 16.2 - type: recall_at_1 value: 63.1 - type: recall_at_10 value: 86.8 - type: recall_at_100 value: 98.1 - type: recall_at_1000 value: 100.0 - type: recall_at_3 value: 76.5 - type: recall_at_5 value: 81.0 - type: main_score value: 74.665 task: type: Retrieval - dataset: config: default name: MTEB MultilingualSentiment (default) revision: None split: validation type: C-MTEB/MultilingualSentiment-classification metrics: - type: accuracy value: 75.98 - type: accuracy_stderr value: 0.8634813257969153 - type: f1 value: 75.98312901227456 - type: f1_stderr value: 0.9813231777702479 - type: main_score value: 75.98 task: type: Classification - dataset: config: default name: MTEB Ocnli (default) revision: None split: validation type: C-MTEB/OCNLI metrics: - type: cos_sim_accuracy value: 80.02165674066053 - type: cos_sim_accuracy_threshold value: 84.70024466514587 - type: cos_sim_ap value: 84.5948682253982 - type: cos_sim_f1 value: 80.84291187739463 - type: cos_sim_f1_threshold value: 82.62853622436523 - type: cos_sim_precision value: 73.97020157756354 - type: cos_sim_recall value: 89.1235480464625 - type: dot_accuracy value: 71.52138603140227 - type: dot_accuracy_threshold value: 84206.94580078125 - type: dot_ap value: 77.69986172282461 - type: dot_f1 value: 74.76467951591216 - type: dot_f1_threshold value: 78842.08984375 - type: dot_precision value: 64.95327102803739 - type: dot_recall value: 88.0675818373812 - type: euclidean_accuracy value: 76.01515971846237 - type: euclidean_accuracy_threshold value: 1818.9674377441406 - type: euclidean_ap value: 80.84369691331835 - type: euclidean_f1 value: 78.08988764044943 - type: euclidean_f1_threshold value: 1922.1363067626953 - type: euclidean_precision value: 70.14297729184187 - type: euclidean_recall value: 88.0675818373812 - type: manhattan_accuracy value: 76.12344342176502 - type: manhattan_accuracy_threshold value: 61934.478759765625 - type: manhattan_ap value: 80.8051823205177 - type: manhattan_f1 value: 78.21596244131456 - type: manhattan_f1_threshold value: 64840.447998046875 - type: manhattan_precision value: 70.41420118343196 - type: manhattan_recall value: 87.96198521647307 - type: max_accuracy value: 80.02165674066053 - type: max_ap value: 84.5948682253982 - type: max_f1 value: 80.84291187739463 task: type: PairClassification - dataset: config: default name: MTEB OnlineShopping (default) revision: None split: test type: C-MTEB/OnlineShopping-classification metrics: - type: accuracy value: 93.63 - type: accuracy_stderr value: 0.7253275122315392 - type: ap value: 91.66092551327398 - type: ap_stderr value: 0.9661774073521741 - type: f1 value: 93.61696896914624 - type: f1_stderr value: 0.7232416235078093 - type: main_score value: 93.63 task: type: Classification - dataset: config: default name: MTEB PAWSX (default) revision: None split: test type: C-MTEB/PAWSX metrics: - type: cosine_pearson value: 27.420084312732477 - type: cosine_spearman value: 36.615019324915316 - type: manhattan_pearson value: 35.38814491527626 - type: manhattan_spearman value: 35.989020517540105 - type: euclidean_pearson value: 35.322828019800475 - type: euclidean_spearman value: 35.93118948093057 - type: main_score value: 36.615019324915316 task: type: STS - dataset: config: default name: MTEB QBQTC (default) revision: None split: test type: C-MTEB/QBQTC metrics: - type: cosine_pearson value: 36.51779732355864 - type: cosine_spearman value: 38.35615142712016 - type: manhattan_pearson value: 31.00096996824444 - type: manhattan_spearman value: 35.22782463612116 - type: euclidean_pearson value: 31.04604995563808 - type: euclidean_spearman value: 35.271420992011485 - type: main_score value: 38.35615142712016 task: type: STS - dataset: config: zh name: MTEB STS22 (zh) revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80 split: test type: mteb/sts22-crosslingual-sts metrics: - type: cosine_pearson value: 60.76376961662733 - type: cosine_spearman value: 65.93112312064913 - type: manhattan_pearson value: 60.18998639945854 - type: manhattan_spearman value: 64.37697612695015 - type: euclidean_pearson value: 60.287759656277814 - type: euclidean_spearman value: 64.37685757691955 - type: main_score value: 65.93112312064913 task: type: STS - dataset: config: default name: MTEB STSB (default) revision: None split: test type: C-MTEB/STSB metrics: - type: cosine_pearson value: 79.6320389543562 - type: cosine_spearman value: 81.9230633773663 - type: manhattan_pearson value: 80.20746913195181 - type: manhattan_spearman value: 80.43150657863002 - type: euclidean_pearson value: 80.1796408157508 - type: euclidean_spearman value: 80.42930201788549 - type: main_score value: 81.9230633773663 task: type: STS - dataset: config: default name: MTEB T2Reranking (default) revision: None split: dev type: C-MTEB/T2Reranking metrics: - type: map value: 66.67836204644267 - type: mrr value: 76.1707222383424 - type: main_score value: 66.67836204644267 task: type: Reranking - dataset: config: default name: MTEB T2Retrieval (default) revision: None split: dev type: C-MTEB/T2Retrieval metrics: - type: map_at_1 value: 28.015 - type: map_at_10 value: 78.281 - type: map_at_100 value: 81.89699999999999 - type: map_at_1000 value: 81.95599999999999 - type: map_at_3 value: 55.117000000000004 - type: map_at_5 value: 67.647 - type: mrr_at_1 value: 90.496 - type: mrr_at_10 value: 93.132 - type: mrr_at_100 value: 93.207 - type: mrr_at_1000 value: 93.209 - type: mrr_at_3 value: 92.714 - type: mrr_at_5 value: 93.0 - type: ndcg_at_1 value: 90.496 - type: ndcg_at_10 value: 85.71600000000001 - type: ndcg_at_100 value: 89.164 - type: ndcg_at_1000 value: 89.71000000000001 - type: ndcg_at_3 value: 86.876 - type: ndcg_at_5 value: 85.607 - type: precision_at_1 value: 90.496 - type: precision_at_10 value: 42.398 - type: precision_at_100 value: 5.031 - type: precision_at_1000 value: 0.516 - type: precision_at_3 value: 75.729 - type: precision_at_5 value: 63.522 - type: recall_at_1 value: 28.015 - type: recall_at_10 value: 84.83000000000001 - type: recall_at_100 value: 95.964 - type: recall_at_1000 value: 98.67399999999999 - type: recall_at_3 value: 56.898 - type: recall_at_5 value: 71.163 - type: main_score value: 85.71600000000001 task: type: Retrieval - dataset: config: default name: MTEB TNews (default) revision: None split: validation type: C-MTEB/TNews-classification metrics: - type: accuracy value: 51.702999999999996 - type: accuracy_stderr value: 0.8183526134863877 - type: f1 value: 50.35330734766769 - type: f1_stderr value: 0.740275098366631 - type: main_score value: 51.702999999999996 task: type: Classification - dataset: config: default name: MTEB ThuNewsClusteringP2P (default) revision: None split: test type: C-MTEB/ThuNewsClusteringP2P metrics: - type: v_measure value: 72.78709391223538 - type: v_measure_std value: 1.5927130767880417 - type: main_score value: 72.78709391223538 task: type: Clustering - dataset: config: default name: MTEB ThuNewsClusteringS2S (default) revision: None split: test type: C-MTEB/ThuNewsClusteringS2S metrics: - type: v_measure value: 66.80392174700211 - type: v_measure_std value: 1.845756306548485 - type: main_score value: 66.80392174700211 task: type: Clustering - dataset: config: default name: MTEB VideoRetrieval (default) revision: None split: dev type: C-MTEB/VideoRetrieval metrics: - type: map_at_1 value: 65.5 - type: map_at_10 value: 75.38 - type: map_at_100 value: 75.756 - type: map_at_1000 value: 75.75800000000001 - type: map_at_3 value: 73.8 - type: map_at_5 value: 74.895 - type: mrr_at_1 value: 65.5 - type: mrr_at_10 value: 75.38 - type: mrr_at_100 value: 75.756 - type: mrr_at_1000 value: 75.75800000000001 - type: mrr_at_3 value: 73.8 - type: mrr_at_5 value: 74.895 - type: ndcg_at_1 value: 65.5 - type: ndcg_at_10 value: 79.572 - type: ndcg_at_100 value: 81.17699999999999 - type: ndcg_at_1000 value: 81.227 - type: ndcg_at_3 value: 76.44999999999999 - type: ndcg_at_5 value: 78.404 - type: precision_at_1 value: 65.5 - type: precision_at_10 value: 9.24 - type: precision_at_100 value: 0.9939999999999999 - type: precision_at_1000 value: 0.1 - type: precision_at_3 value: 28.033 - type: precision_at_5 value: 17.76 - type: recall_at_1 value: 65.5 - type: recall_at_10 value: 92.4 - type: recall_at_100 value: 99.4 - type: recall_at_1000 value: 99.8 - type: recall_at_3 value: 84.1 - type: recall_at_5 value: 88.8 - type: main_score value: 79.572 task: type: Retrieval - dataset: config: default name: MTEB Waimai (default) revision: None split: test type: C-MTEB/waimai-classification metrics: - type: accuracy value: 88.70000000000002 - type: accuracy_stderr value: 1.1713240371477067 - type: ap value: 73.95357766936226 - type: ap_stderr value: 2.3258932220157638 - type: f1 value: 87.27541455081986 - type: f1_stderr value: 1.185968184225313 - type: main_score value: 88.70000000000002 task: type: Classification tags: - mteb --- ## Yuan-embedding-1.0 Yuan-embedding-1.0 是专门为中文文本检索任务设计的嵌入模型。 在xiaobu模型结构(bert-large结构)基础上, 采用全新的数据集构建、生成与清洗方法, 结合二阶段微调实现Retrieval任务的精度领先(Hugging Face C-MTEB榜单 [1])。 其中, 正负例样本采用源2.0-M32(Yuan2.0-M32 [2])大模型进行生成。主要工作如下： - 在Hard negative sampling中，使用Rerank模型(bge-reranker-large [3])进行数据排序筛选 - 通过(Yuan2.0-M32大模型)迭代生成新query、corpus - 采用MRL方法进行模型微调训练 ## Usage ```bash pip install -U sentence-transformers==3.1.1 ``` 使用示例： ```python from sentence_transformers import SentenceTransformer model = SentenceTransformer("IEIYuan/Yuan-embedding-1.0") sentences = [ "这是一个样例-1", "这是一个样例-2", ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities) ``` ## Reference 1. https://huggingface.co./spaces/mteb/leaderboard 2. https://huggingface.co./IEITYuan/Yuan2-M32 3. https://huggingface.co./BAAI/bge-reranker-large