--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers - sentence-embedding - mteb model-index: - name: bilingual-embedding-large-8k results: - task: type: Clustering dataset: type: lyon-nlp/alloprof name: MTEB AlloProfClusteringP2P config: default split: test revision: 392ba3f5bcc8c51f578786c1fc3dae648662cb9b metrics: - type: v_measure value: 55.52298673909706 - type: v_measures value: [0.5198748380785404, 0.5562521099012603, 0.5322986254464575, 0.5722250987615152, 0.532932258758668] - task: type: Clustering dataset: type: lyon-nlp/alloprof name: MTEB AlloProfClusteringS2S config: default split: test revision: 392ba3f5bcc8c51f578786c1fc3dae648662cb9b metrics: - type: v_measure value: 35.802733348353094 - type: v_measures value: [0.37359796790048144, 0.36376421464272285, 0.37524966704915225, 0.3749296797757371, 0.36673700158106576] - task: type: Reranking dataset: type: lyon-nlp/mteb-fr-reranking-alloprof-s2p name: MTEB AlloprofReranking config: default split: test revision: 65393d0d7a08a10b4e348135e824f385d420b0fd metrics: - type: map value: 73.10088493122083 - type: mrr value: 74.33452929243086 - type: nAUC_map_diff1 value: 56.63750231223696 - type: nAUC_map_max value: 27.066268470355492 - type: nAUC_mrr_diff1 value: 55.33487252773409 - type: nAUC_mrr_max value: 27.328424865584367 - task: type: Retrieval dataset: type: lyon-nlp/alloprof name: MTEB AlloprofRetrieval config: default split: test revision: fcf295ea64c750f41fadbaa37b9b861558e1bfbd metrics: - type: map_at_1 value: 28.282 - type: map_at_10 value: 38.805 - type: map_at_100 value: 39.804 - type: map_at_1000 value: 39.859 - type: map_at_20 value: 39.428999999999995 - type: map_at_3 value: 35.838 - type: map_at_5 value: 37.537 - type: mrr_at_1 value: 28.281519861830745 - type: mrr_at_10 value: 38.805171752063 - type: mrr_at_100 value: 39.80444636265999 - type: mrr_at_1000 value: 39.858722309770464 - type: mrr_at_20 value: 39.42867574383368 - type: mrr_at_3 value: 35.83765112262528 - type: mrr_at_5 value: 37.53670120898111 - type: nauc_map_at_1000_diff1 value: 42.44333601195352 - type: nauc_map_at_1000_max value: 41.88361927698375 - type: nauc_map_at_100_diff1 value: 42.42746715522874 - type: nauc_map_at_100_max value: 41.913701611267015 - type: nauc_map_at_10_diff1 value: 42.25094726032311 - type: nauc_map_at_10_max value: 41.76772459035808 - type: nauc_map_at_1_diff1 value: 48.03355307282109 - type: nauc_map_at_1_max value: 38.73226070718987 - type: nauc_map_at_20_diff1 value: 42.3550770875435 - type: nauc_map_at_20_max value: 41.89957212013687 - type: nauc_map_at_3_diff1 value: 42.88695727955848 - type: nauc_map_at_3_max value: 40.81262402287836 - type: nauc_map_at_5_diff1 value: 42.34041989483334 - type: nauc_map_at_5_max value: 41.36458206255729 - type: nauc_mrr_at_1000_diff1 value: 42.44333601195352 - type: nauc_mrr_at_1000_max value: 41.88361927698375 - type: nauc_mrr_at_100_diff1 value: 42.42746715522874 - type: nauc_mrr_at_100_max value: 41.913701611267015 - type: nauc_mrr_at_10_diff1 value: 42.25094726032311 - type: nauc_mrr_at_10_max value: 41.76772459035808 - type: nauc_mrr_at_1_diff1 value: 48.03355307282109 - type: nauc_mrr_at_1_max value: 38.73226070718987 - type: nauc_mrr_at_20_diff1 value: 42.3550770875435 - type: nauc_mrr_at_20_max value: 41.89957212013687 - type: nauc_mrr_at_3_diff1 value: 42.88695727955848 - type: nauc_mrr_at_3_max value: 40.81262402287836 - type: nauc_mrr_at_5_diff1 value: 42.34041989483334 - type: nauc_mrr_at_5_max value: 41.36458206255729 - type: nauc_ndcg_at_1000_diff1 value: 41.35830258715452 - type: nauc_ndcg_at_1000_max value: 43.2765379475269 - type: nauc_ndcg_at_100_diff1 value: 40.95047094384412 - type: nauc_ndcg_at_100_max value: 44.293436979483594 - type: nauc_ndcg_at_10_diff1 value: 40.0359339979518 - type: nauc_ndcg_at_10_max value: 43.390909520520076 - type: nauc_ndcg_at_1_diff1 value: 48.03355307282109 - type: nauc_ndcg_at_1_max value: 38.73226070718987 - type: nauc_ndcg_at_20_diff1 value: 40.35056898575259 - type: nauc_ndcg_at_20_max value: 43.991764985610345 - type: nauc_ndcg_at_3_diff1 value: 41.40129960980627 - type: nauc_ndcg_at_3_max value: 41.41104483378663 - type: nauc_ndcg_at_5_diff1 value: 40.36007384476364 - type: nauc_ndcg_at_5_max value: 42.383481303106414 - type: nauc_precision_at_1000_diff1 value: 34.36657255351115 - type: nauc_precision_at_1000_max value: 74.91976431868189 - type: nauc_precision_at_100_diff1 value: 33.55702830592739 - type: nauc_precision_at_100_max value: 65.71347416493107 - type: nauc_precision_at_10_diff1 value: 32.521448032129 - type: nauc_precision_at_10_max value: 49.19930788953475 - type: nauc_precision_at_1_diff1 value: 48.03355307282109 - type: nauc_precision_at_1_max value: 38.73226070718987 - type: nauc_precision_at_20_diff1 value: 32.57892299703891 - type: nauc_precision_at_20_max value: 53.45967162017302 - type: nauc_precision_at_3_diff1 value: 37.249551957650795 - type: nauc_precision_at_3_max value: 43.08267504682664 - type: nauc_precision_at_5_diff1 value: 34.44393985129692 - type: nauc_precision_at_5_max value: 45.460096642832646 - type: nauc_recall_at_1000_diff1 value: 34.36657255350965 - type: nauc_recall_at_1000_max value: 74.91976431868211 - type: nauc_recall_at_100_diff1 value: 33.55702830592741 - type: nauc_recall_at_100_max value: 65.71347416493116 - type: nauc_recall_at_10_diff1 value: 32.52144803212901 - type: nauc_recall_at_10_max value: 49.199307889534715 - type: nauc_recall_at_1_diff1 value: 48.03355307282109 - type: nauc_recall_at_1_max value: 38.73226070718987 - type: nauc_recall_at_20_diff1 value: 32.57892299703892 - type: nauc_recall_at_20_max value: 53.45967162017302 - type: nauc_recall_at_3_diff1 value: 37.249551957650766 - type: nauc_recall_at_3_max value: 43.082675046826644 - type: nauc_recall_at_5_diff1 value: 34.4439398512969 - type: nauc_recall_at_5_max value: 45.460096642832674 - type: ndcg_at_1 value: 28.282 - type: ndcg_at_10 value: 44.421 - type: ndcg_at_100 value: 49.447 - type: ndcg_at_1000 value: 50.981 - type: ndcg_at_20 value: 46.671 - type: ndcg_at_3 value: 38.289 - type: ndcg_at_5 value: 41.349999999999994 - type: precision_at_1 value: 28.282 - type: precision_at_10 value: 6.231 - type: precision_at_100 value: 0.8619999999999999 - type: precision_at_1000 value: 0.098 - type: precision_at_20 value: 3.558 - type: precision_at_3 value: 15.126999999999999 - type: precision_at_5 value: 10.561 - type: recall_at_1 value: 28.282 - type: recall_at_10 value: 62.306 - type: recall_at_100 value: 86.226 - type: recall_at_1000 value: 98.489 - type: recall_at_20 value: 71.15700000000001 - type: recall_at_3 value: 45.379999999999995 - type: recall_at_5 value: 52.807 - task: type: Classification dataset: type: mteb/amazon_reviews_multi name: MTEB AmazonReviewsClassification (fr) config: fr split: test revision: 1399c76144fd37290681b995c656ef9b2e06e26d metrics: - type: accuracy value: 44.10999999999999 - type: f1 value: 42.00584553745547 - type: f1_weighted value: 42.005845537455485 - task: type: Retrieval dataset: type: maastrichtlawtech/bsard name: MTEB BSARDRetrieval config: default split: test revision: 5effa1b9b5fa3b0f9e12523e6e43e5f86a6e6d59 metrics: - type: map_at_1 value: 4.955 - type: map_at_10 value: 9.103 - type: map_at_100 value: 9.998999999999999 - type: map_at_1000 value: 10.136000000000001 - type: map_at_20 value: 9.554 - type: map_at_3 value: 7.4319999999999995 - type: map_at_5 value: 7.95 - type: mrr_at_1 value: 4.954954954954955 - type: mrr_at_10 value: 9.102852852852852 - type: mrr_at_100 value: 9.999215850941926 - type: mrr_at_1000 value: 10.13616308946331 - type: mrr_at_20 value: 9.554402632963003 - type: mrr_at_3 value: 7.432432432432434 - type: mrr_at_5 value: 7.950450450450451 - type: nauc_map_at_1000_diff1 value: 14.655819915811785 - type: nauc_map_at_1000_max value: 9.188182207979008 - type: nauc_map_at_100_diff1 value: 14.517637755979687 - type: nauc_map_at_100_max value: 9.060725563022503 - type: nauc_map_at_10_diff1 value: 15.776144582905358 - type: nauc_map_at_10_max value: 9.448668398689462 - type: nauc_map_at_1_diff1 value: 19.10921794840591 - type: nauc_map_at_1_max value: 4.060331068810239 - type: nauc_map_at_20_diff1 value: 15.061809327427353 - type: nauc_map_at_20_max value: 9.085953657690329 - type: nauc_map_at_3_diff1 value: 18.42793018906856 - type: nauc_map_at_3_max value: 10.10140103912974 - type: nauc_map_at_5_diff1 value: 17.407972669931233 - type: nauc_map_at_5_max value: 10.064885264376228 - type: nauc_mrr_at_1000_diff1 value: 14.655819915811785 - type: nauc_mrr_at_1000_max value: 9.188182207979008 - type: nauc_mrr_at_100_diff1 value: 14.517637755979687 - type: nauc_mrr_at_100_max value: 9.060725563022503 - type: nauc_mrr_at_10_diff1 value: 15.776144582905358 - type: nauc_mrr_at_10_max value: 9.448668398689462 - type: nauc_mrr_at_1_diff1 value: 19.10921794840591 - type: nauc_mrr_at_1_max value: 4.060331068810239 - type: nauc_mrr_at_20_diff1 value: 15.061809327427353 - type: nauc_mrr_at_20_max value: 9.085953657690329 - type: nauc_mrr_at_3_diff1 value: 18.42793018906856 - type: nauc_mrr_at_3_max value: 10.10140103912974 - type: nauc_mrr_at_5_diff1 value: 17.407972669931233 - type: nauc_mrr_at_5_max value: 10.064885264376228 - type: nauc_ndcg_at_1000_diff1 value: 11.940580725648152 - type: nauc_ndcg_at_1000_max value: 11.004283166102807 - type: nauc_ndcg_at_100_diff1 value: 10.009680762933215 - type: nauc_ndcg_at_100_max value: 8.444186642393188 - type: nauc_ndcg_at_10_diff1 value: 14.423251037136561 - type: nauc_ndcg_at_10_max value: 10.614014795363303 - type: nauc_ndcg_at_1_diff1 value: 19.10921794840591 - type: nauc_ndcg_at_1_max value: 4.060331068810239 - type: nauc_ndcg_at_20_diff1 value: 12.486198272876521 - type: nauc_ndcg_at_20_max value: 9.550225653436467 - type: nauc_ndcg_at_3_diff1 value: 18.813915768129757 - type: nauc_ndcg_at_3_max value: 11.865670858870484 - type: nauc_ndcg_at_5_diff1 value: 17.01715479783127 - type: nauc_ndcg_at_5_max value: 11.523181173967899 - type: nauc_precision_at_1000_diff1 value: 6.162580085242911 - type: nauc_precision_at_1000_max value: 21.74545120171883 - type: nauc_precision_at_100_diff1 value: 1.4492186570094137 - type: nauc_precision_at_100_max value: 5.320582161712451 - type: nauc_precision_at_10_diff1 value: 12.199838986983115 - type: nauc_precision_at_10_max value: 12.409471572004998 - type: nauc_precision_at_1_diff1 value: 19.10921794840591 - type: nauc_precision_at_1_max value: 4.060331068810239 - type: nauc_precision_at_20_diff1 value: 8.089525252638769 - type: nauc_precision_at_20_max value: 9.829600854870332 - type: nauc_precision_at_3_diff1 value: 19.71630962813128 - type: nauc_precision_at_3_max value: 15.560242379569136 - type: nauc_precision_at_5_diff1 value: 16.151579517326258 - type: nauc_precision_at_5_max value: 14.225120177799683 - type: nauc_recall_at_1000_diff1 value: 6.1625800852429595 - type: nauc_recall_at_1000_max value: 21.745451201718687 - type: nauc_recall_at_100_diff1 value: 1.4492186570093863 - type: nauc_recall_at_100_max value: 5.320582161712405 - type: nauc_recall_at_10_diff1 value: 12.199838986983083 - type: nauc_recall_at_10_max value: 12.409471572004962 - type: nauc_recall_at_1_diff1 value: 19.10921794840591 - type: nauc_recall_at_1_max value: 4.060331068810239 - type: nauc_recall_at_20_diff1 value: 8.089525252638692 - type: nauc_recall_at_20_max value: 9.829600854870273 - type: nauc_recall_at_3_diff1 value: 19.716309628131278 - type: nauc_recall_at_3_max value: 15.560242379569129 - type: nauc_recall_at_5_diff1 value: 16.151579517326265 - type: nauc_recall_at_5_max value: 14.225120177799697 - type: ndcg_at_1 value: 4.955 - type: ndcg_at_10 value: 12.005 - type: ndcg_at_100 value: 17.238 - type: ndcg_at_1000 value: 21.287 - type: ndcg_at_20 value: 13.691999999999998 - type: ndcg_at_3 value: 8.296000000000001 - type: ndcg_at_5 value: 9.225999999999999 - type: precision_at_1 value: 4.955 - type: precision_at_10 value: 2.162 - type: precision_at_100 value: 0.482 - type: precision_at_1000 value: 0.08099999999999999 - type: precision_at_20 value: 1.419 - type: precision_at_3 value: 3.604 - type: precision_at_5 value: 2.613 - type: recall_at_1 value: 4.955 - type: recall_at_10 value: 21.622 - type: recall_at_100 value: 48.198 - type: recall_at_1000 value: 81.081 - type: recall_at_20 value: 28.377999999999997 - type: recall_at_3 value: 10.811 - type: recall_at_5 value: 13.062999999999999 - task: type: Clustering dataset: type: lyon-nlp/clustering-hal-s2s name: MTEB HALClusteringS2S config: default split: test revision: e06ebbbb123f8144bef1a5d18796f3dec9ae2915 metrics: - type: v_measure value: 23.137623974622365 - type: v_measures value: [0.2802068838665942, 0.2565274984774815, 0.25245022445056786, 0.22595460950575297, 0.20177741591393913] - task: type: Clustering dataset: type: reciTAL/mlsum name: MTEB MLSUMClusteringP2P config: default split: test revision: b5d54f8f3b61ae17845046286940f03c6bc79bc7 metrics: - type: v_measure value: 40.31003279146524 - type: v_measures value: [0.3943651322813771, 0.4189000344922205, 0.4101443880670743, 0.3832149080991847, 0.37602613534689566] - task: type: Clustering dataset: type: reciTAL/mlsum name: MTEB MLSUMClusteringS2S config: default split: test revision: b5d54f8f3b61ae17845046286940f03c6bc79bc7 metrics: - type: v_measure value: 40.04524841336757 - type: v_measures value: [0.39835449199860185, 0.405905613221237, 0.40326782414397255, 0.40882879348632284, 0.3683302592759367] - task: type: Classification dataset: type: mteb/mtop_domain name: MTEB MTOPDomainClassification (fr) config: fr split: test revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf metrics: - type: accuracy value: 87.82023175696837 - type: f1 value: 87.58287510797385 - type: f1_weighted value: 87.75645870762435 - task: type: Classification dataset: type: mteb/mtop_intent name: MTEB MTOPIntentClassification (fr) config: fr split: test revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba metrics: - type: accuracy value: 58.628249295333546 - type: f1 value: 42.22070573172825 - type: f1_weighted value: 60.62087995743649 - task: type: Classification dataset: type: mteb/masakhanews name: MTEB MasakhaNEWSClassification (fra) config: fra split: test revision: 18193f187b92da67168c655c9973a165ed9593dd metrics: - type: accuracy value: 69.81042654028435 - type: f1 value: 66.05811881796396 - type: f1_weighted value: 70.34901566149948 - task: type: Clustering dataset: type: masakhane/masakhanews name: MTEB MasakhaNEWSClusteringP2P (fra) config: fra split: test revision: 8ccc72e69e65f40c70e117d8b3c08306bb788b60 metrics: - type: v_measure value: 45.02712178986078 - type: v_measures value: [1.0, 0.23955793240111928, 0.7158920010774062, 0.036391635653837, 0.25951452036067674] - task: type: Clustering dataset: type: masakhane/masakhanews name: MTEB MasakhaNEWSClusteringS2S (fra) config: fra split: test revision: 8ccc72e69e65f40c70e117d8b3c08306bb788b60 metrics: - type: v_measure value: 30.38607254306223 - type: v_measures value: [1.0, 0.01936507478006705, 0.19876372667844472, 0.17182595867380823, 0.12934886702079137] - task: type: Classification dataset: type: mteb/amazon_massive_intent name: MTEB MassiveIntentClassification (fr) config: fr split: test revision: 4672e20407010da34463acc759c162ca9734bca6 metrics: - type: accuracy value: 66.13651647612645 - type: f1 value: 64.42898347709598 - type: f1_weighted value: 65.01442547020224 - task: type: Classification dataset: type: mteb/amazon_massive_scenario name: MTEB MassiveScenarioClassification (fr) config: fr split: test revision: fad2c6e8459f9e1c45d9315f4953d921437d70f8 metrics: - type: accuracy value: 72.73705447209144 - type: f1 value: 72.09285609231057 - type: f1_weighted value: 72.34295244611339 - task: type: Retrieval dataset: type: jinaai/mintakaqa name: MTEB MintakaRetrieval (fr) config: fr split: test revision: efa78cc2f74bbcd21eff2261f9e13aebe40b814e metrics: - type: map_at_1 value: 13.677 - type: map_at_10 value: 21.044 - type: map_at_100 value: 22.012 - type: map_at_1000 value: 22.125 - type: map_at_20 value: 21.573999999999998 - type: map_at_3 value: 18.857 - type: map_at_5 value: 19.936999999999998 - type: mrr_at_1 value: 13.677313677313677 - type: mrr_at_10 value: 21.043933543933505 - type: mrr_at_100 value: 22.012160523798318 - type: mrr_at_1000 value: 22.124555014776913 - type: mrr_at_20 value: 21.574199922074904 - type: mrr_at_3 value: 18.857493857493825 - type: mrr_at_5 value: 19.93652743652738 - type: nauc_map_at_1000_diff1 value: 20.63087633823352 - type: nauc_map_at_1000_max value: 31.753246807516362 - type: nauc_map_at_100_diff1 value: 20.602874174259885 - type: nauc_map_at_100_max value: 31.74109681792161 - type: nauc_map_at_10_diff1 value: 20.82028964049537 - type: nauc_map_at_10_max value: 32.082751313883705 - type: nauc_map_at_1_diff1 value: 27.838566854973656 - type: nauc_map_at_1_max value: 32.0217755083183 - type: nauc_map_at_20_diff1 value: 20.685607874578192 - type: nauc_map_at_20_max value: 31.89765440964895 - type: nauc_map_at_3_diff1 value: 22.385335765437958 - type: nauc_map_at_3_max value: 32.47346568889047 - type: nauc_map_at_5_diff1 value: 21.173253596770003 - type: nauc_map_at_5_max value: 32.2528418460596 - type: nauc_mrr_at_1000_diff1 value: 20.63087633823352 - type: nauc_mrr_at_1000_max value: 31.753246807516362 - type: nauc_mrr_at_100_diff1 value: 20.602874174259885 - type: nauc_mrr_at_100_max value: 31.74109681792161 - type: nauc_mrr_at_10_diff1 value: 20.82028964049537 - type: nauc_mrr_at_10_max value: 32.082751313883705 - type: nauc_mrr_at_1_diff1 value: 27.838566854973656 - type: nauc_mrr_at_1_max value: 32.0217755083183 - type: nauc_mrr_at_20_diff1 value: 20.685607874578192 - type: nauc_mrr_at_20_max value: 31.89765440964895 - type: nauc_mrr_at_3_diff1 value: 22.385335765437958 - type: nauc_mrr_at_3_max value: 32.47346568889047 - type: nauc_mrr_at_5_diff1 value: 21.173253596770003 - type: nauc_mrr_at_5_max value: 32.2528418460596 - type: nauc_ndcg_at_1000_diff1 value: 18.08460876388022 - type: nauc_ndcg_at_1000_max value: 30.282810360048217 - type: nauc_ndcg_at_100_diff1 value: 17.119539175602068 - type: nauc_ndcg_at_100_max value: 29.66409825853174 - type: nauc_ndcg_at_10_diff1 value: 18.23254548133648 - type: nauc_ndcg_at_10_max value: 31.52995550586078 - type: nauc_ndcg_at_1_diff1 value: 27.838566854973656 - type: nauc_ndcg_at_1_max value: 32.0217755083183 - type: nauc_ndcg_at_20_diff1 value: 17.769003159911446 - type: nauc_ndcg_at_20_max value: 30.929703630445033 - type: nauc_ndcg_at_3_diff1 value: 20.96979719261237 - type: nauc_ndcg_at_3_max value: 32.363993132409526 - type: nauc_ndcg_at_5_diff1 value: 19.00106027591966 - type: nauc_ndcg_at_5_max value: 31.962682994281664 - type: nauc_precision_at_1000_diff1 value: -0.439767274118902 - type: nauc_precision_at_1000_max value: 12.247737195943136 - type: nauc_precision_at_100_diff1 value: 5.574224743755663 - type: nauc_precision_at_100_max value: 20.625486141114006 - type: nauc_precision_at_10_diff1 value: 12.116438700823444 - type: nauc_precision_at_10_max value: 30.027073824365324 - type: nauc_precision_at_1_diff1 value: 27.838566854973656 - type: nauc_precision_at_1_max value: 32.0217755083183 - type: nauc_precision_at_20_diff1 value: 10.528730914479825 - type: nauc_precision_at_20_max value: 28.101643683820228 - type: nauc_precision_at_3_diff1 value: 17.575083081784413 - type: nauc_precision_at_3_max value: 32.04257948042897 - type: nauc_precision_at_5_diff1 value: 13.87097676219356 - type: nauc_precision_at_5_max value: 31.186621554981798 - type: nauc_recall_at_1000_diff1 value: -0.4397672741187951 - type: nauc_recall_at_1000_max value: 12.247737195943454 - type: nauc_recall_at_100_diff1 value: 5.574224743755691 - type: nauc_recall_at_100_max value: 20.625486141114028 - type: nauc_recall_at_10_diff1 value: 12.116438700823482 - type: nauc_recall_at_10_max value: 30.027073824365335 - type: nauc_recall_at_1_diff1 value: 27.838566854973656 - type: nauc_recall_at_1_max value: 32.0217755083183 - type: nauc_recall_at_20_diff1 value: 10.528730914479794 - type: nauc_recall_at_20_max value: 28.101643683820228 - type: nauc_recall_at_3_diff1 value: 17.57508308178443 - type: nauc_recall_at_3_max value: 32.042579480429 - type: nauc_recall_at_5_diff1 value: 13.870976762193543 - type: nauc_recall_at_5_max value: 31.186621554981787 - type: ndcg_at_1 value: 13.677 - type: ndcg_at_10 value: 25.191000000000003 - type: ndcg_at_100 value: 30.379 - type: ndcg_at_1000 value: 33.961999999999996 - type: ndcg_at_20 value: 27.1 - type: ndcg_at_3 value: 20.546 - type: ndcg_at_5 value: 22.505 - type: precision_at_1 value: 13.677 - type: precision_at_10 value: 3.853 - type: precision_at_100 value: 0.639 - type: precision_at_1000 value: 0.093 - type: precision_at_20 value: 2.3009999999999997 - type: precision_at_3 value: 8.477 - type: precision_at_5 value: 6.0440000000000005 - type: recall_at_1 value: 13.677 - type: recall_at_10 value: 38.534 - type: recall_at_100 value: 63.922999999999995 - type: recall_at_1000 value: 93.407 - type: recall_at_20 value: 46.028000000000006 - type: recall_at_3 value: 25.430000000000003 - type: recall_at_5 value: 30.220999999999997 - task: type: PairClassification dataset: type: GEM/opusparcus name: MTEB OpusparcusPC (fr) config: fr split: test revision: 9e9b1f8ef51616073f47f306f7f47dd91663f86a metrics: - type: cos_sim_accuracy value: 82.9700272479564 - type: cos_sim_ap value: 93.15021785539084 - type: cos_sim_f1 value: 87.97316722568279 - type: cos_sim_precision value: 85.0 - type: cos_sim_recall value: 91.16186693147964 - type: dot_accuracy value: 82.9700272479564 - type: dot_ap value: 93.15021785539084 - type: dot_f1 value: 87.97316722568279 - type: dot_precision value: 85.0 - type: dot_recall value: 91.16186693147964 - type: euclidean_accuracy value: 82.9700272479564 - type: euclidean_ap value: 93.15015081441638 - type: euclidean_f1 value: 87.97316722568279 - type: euclidean_precision value: 85.0 - type: euclidean_recall value: 91.16186693147964 - type: manhattan_accuracy value: 82.56130790190735 - type: manhattan_ap value: 93.14590481820592 - type: manhattan_f1 value: 87.86729857819905 - type: manhattan_precision value: 84.04351767905711 - type: manhattan_recall value: 92.05561072492552 - type: max_accuracy value: 82.9700272479564 - type: max_ap value: 93.15021785539084 - type: max_f1 value: 87.97316722568279 - task: type: PairClassification dataset: type: google-research-datasets/paws-x name: MTEB PawsX (fr) config: fr split: test revision: 8a04d940a42cd40658986fdd8e3da561533a3646 metrics: - type: cos_sim_accuracy value: 64.14999999999999 - type: cos_sim_ap value: 63.43794001840604 - type: cos_sim_f1 value: 62.59187620889749 - type: cos_sim_precision value: 48.097502972651604 - type: cos_sim_recall value: 89.59025470653378 - type: dot_accuracy value: 64.14999999999999 - type: dot_ap value: 63.52400235031554 - type: dot_f1 value: 62.59187620889749 - type: dot_precision value: 48.097502972651604 - type: dot_recall value: 89.59025470653378 - type: euclidean_accuracy value: 64.14999999999999 - type: euclidean_ap value: 63.43794001840604 - type: euclidean_f1 value: 62.59187620889749 - type: euclidean_precision value: 48.097502972651604 - type: euclidean_recall value: 89.59025470653378 - type: manhattan_accuracy value: 64.2 - type: manhattan_ap value: 63.46163243480347 - type: manhattan_f1 value: 62.540021344717175 - type: manhattan_precision value: 46.069182389937104 - type: manhattan_recall value: 97.34219269102991 - type: max_accuracy value: 64.2 - type: max_ap value: 63.52400235031554 - type: max_f1 value: 62.59187620889749 - task: type: STS dataset: type: Lajavaness/SICK-fr name: MTEB SICKFr config: default split: test revision: e077ab4cf4774a1e36d86d593b150422fafd8e8a metrics: - type: cos_sim_pearson value: 85.12347242597652 - type: cos_sim_spearman value: 79.80580538857501 - type: euclidean_pearson value: 82.03127787921382 - type: euclidean_spearman value: 79.80580538857501 - type: manhattan_pearson value: 82.02795155003601 - type: manhattan_spearman value: 79.7808784011127 - task: type: STS dataset: type: mteb/sts22-crosslingual-sts name: MTEB STS22 (fr) config: fr split: test revision: de9d86b3b84231dc21f76c7b7af1f28e2f57f6e3 metrics: - type: cos_sim_pearson value: 82.34462624659417 - type: cos_sim_spearman value: 82.83867899462683 - type: euclidean_pearson value: 80.00679113308384 - type: euclidean_spearman value: 82.83867899462683 - type: manhattan_pearson value: 79.97582730301362 - type: manhattan_spearman value: 82.95718926500541 - task: type: STS dataset: type: mteb/stsb_multi_mt name: MTEB STSBenchmarkMultilingualSTS (fr) config: fr split: test revision: 29afa2569dcedaaa2fe6a3dcfebab33d28b82e8c metrics: - type: cos_sim_pearson value: 86.0897698618904 - type: cos_sim_spearman value: 86.58814894229229 - type: euclidean_pearson value: 85.53992615842806 - type: euclidean_spearman value: 86.58814894229229 - type: manhattan_pearson value: 85.4985023034774 - type: manhattan_spearman value: 86.50239881298486 - task: type: Summarization dataset: type: lyon-nlp/summarization-summeval-fr-p2p name: MTEB SummEvalFr config: default split: test revision: b385812de6a9577b6f4d0f88c6a6e35395a94054 metrics: - type: cos_sim_pearson value: 30.458145110977753 - type: cos_sim_spearman value: 31.624715940109265 - type: dot_pearson value: 30.458145236239915 - type: dot_spearman value: 31.624715940109265 - task: type: Reranking dataset: type: lyon-nlp/mteb-fr-reranking-syntec-s2p name: MTEB SyntecReranking config: default split: test revision: daf0863838cd9e3ba50544cdce3ac2b338a1b0ad metrics: - type: map value: 87.60277777777777 - type: mrr value: 87.60277777777777 - type: nAUC_map_diff1 value: 63.877496103879814 - type: nAUC_map_max value: -4.8943605546581725 - type: nAUC_mrr_diff1 value: 63.877496103879814 - type: nAUC_mrr_max value: -4.8943605546581725 - task: type: Retrieval dataset: type: lyon-nlp/mteb-fr-retrieval-syntec-s2p name: MTEB SyntecRetrieval config: default split: test revision: 19661ccdca4dfc2d15122d776b61685f48c68ca9 metrics: - type: map_at_1 value: 67.0 - type: map_at_10 value: 78.47800000000001 - type: map_at_100 value: 78.616 - type: map_at_1000 value: 78.616 - type: map_at_20 value: 78.52799999999999 - type: map_at_3 value: 77.833 - type: map_at_5 value: 78.033 - type: mrr_at_1 value: 67.0 - type: mrr_at_10 value: 78.47777777777777 - type: mrr_at_100 value: 78.61609758846066 - type: mrr_at_1000 value: 78.61609758846066 - type: mrr_at_20 value: 78.52777777777777 - type: mrr_at_3 value: 77.83333333333331 - type: mrr_at_5 value: 78.03333333333333 - type: nauc_map_at_1000_diff1 value: 54.76919250379753 - type: nauc_map_at_1000_max value: 24.03294042759147 - type: nauc_map_at_100_diff1 value: 54.76919250379753 - type: nauc_map_at_100_max value: 24.03294042759147 - type: nauc_map_at_10_diff1 value: 54.781660658782506 - type: nauc_map_at_10_max value: 24.45332707633837 - type: nauc_map_at_1_diff1 value: 54.48189466912695 - type: nauc_map_at_1_max value: 17.502666282947597 - type: nauc_map_at_20_diff1 value: 54.69518355408933 - type: nauc_map_at_20_max value: 24.285263763068183 - type: nauc_map_at_3_diff1 value: 54.98928575752318 - type: nauc_map_at_3_max value: 25.252117626643916 - type: nauc_map_at_5_diff1 value: 54.51750311747391 - type: nauc_map_at_5_max value: 25.141479081321766 - type: nauc_mrr_at_1000_diff1 value: 54.76919250379753 - type: nauc_mrr_at_1000_max value: 24.03294042759147 - type: nauc_mrr_at_100_diff1 value: 54.76919250379753 - type: nauc_mrr_at_100_max value: 24.03294042759147 - type: nauc_mrr_at_10_diff1 value: 54.781660658782506 - type: nauc_mrr_at_10_max value: 24.45332707633837 - type: nauc_mrr_at_1_diff1 value: 54.48189466912695 - type: nauc_mrr_at_1_max value: 17.502666282947597 - type: nauc_mrr_at_20_diff1 value: 54.69518355408933 - type: nauc_mrr_at_20_max value: 24.285263763068183 - type: nauc_mrr_at_3_diff1 value: 54.98928575752318 - type: nauc_mrr_at_3_max value: 25.252117626643916 - type: nauc_mrr_at_5_diff1 value: 54.51750311747391 - type: nauc_mrr_at_5_max value: 25.141479081321766 - type: nauc_ndcg_at_1000_diff1 value: 54.411394691026196 - type: nauc_ndcg_at_1000_max value: 25.003182969921014 - type: nauc_ndcg_at_100_diff1 value: 54.411394691026196 - type: nauc_ndcg_at_100_max value: 25.003182969921014 - type: nauc_ndcg_at_10_diff1 value: 53.97509194326736 - type: nauc_ndcg_at_10_max value: 27.51736442048005 - type: nauc_ndcg_at_1_diff1 value: 54.48189466912695 - type: nauc_ndcg_at_1_max value: 17.502666282947597 - type: nauc_ndcg_at_20_diff1 value: 53.46713794714154 - type: nauc_ndcg_at_20_max value: 26.601577957753005 - type: nauc_ndcg_at_3_diff1 value: 54.521393171396525 - type: nauc_ndcg_at_3_max value: 29.07380139412928 - type: nauc_ndcg_at_5_diff1 value: 53.42255297135452 - type: nauc_ndcg_at_5_max value: 28.91110004742623 - type: nauc_precision_at_1000_diff1 value: nan - type: nauc_precision_at_1000_max value: nan - type: nauc_precision_at_100_diff1 value: nan - type: nauc_precision_at_100_max value: nan - type: nauc_precision_at_10_diff1 value: 41.59663865546228 - type: nauc_precision_at_10_max value: 67.44864612511667 - type: nauc_precision_at_1_diff1 value: 54.48189466912695 - type: nauc_precision_at_1_max value: 17.502666282947597 - type: nauc_precision_at_20_diff1 value: 26.486150015561265 - type: nauc_precision_at_20_max value: 60.95549330843449 - type: nauc_precision_at_3_diff1 value: 50.78781512605074 - type: nauc_precision_at_3_max value: 55.48552754435131 - type: nauc_precision_at_5_diff1 value: 43.75750300120062 - type: nauc_precision_at_5_max value: 58.29665199413101 - type: nauc_recall_at_1000_diff1 value: nan - type: nauc_recall_at_1000_max value: nan - type: nauc_recall_at_100_diff1 value: nan - type: nauc_recall_at_100_max value: nan - type: nauc_recall_at_10_diff1 value: 41.59663865546242 - type: nauc_recall_at_10_max value: 67.44864612511677 - type: nauc_recall_at_1_diff1 value: 54.48189466912695 - type: nauc_recall_at_1_max value: 17.502666282947597 - type: nauc_recall_at_20_diff1 value: 26.486150015561737 - type: nauc_recall_at_20_max value: 60.95549330843472 - type: nauc_recall_at_3_diff1 value: 50.787815126050376 - type: nauc_recall_at_3_max value: 55.48552754435111 - type: nauc_recall_at_5_diff1 value: 43.75750300120054 - type: nauc_recall_at_5_max value: 58.29665199413113 - type: ndcg_at_1 value: 67.0 - type: ndcg_at_10 value: 82.864 - type: ndcg_at_100 value: 83.672 - type: ndcg_at_1000 value: 83.672 - type: ndcg_at_20 value: 83.092 - type: ndcg_at_3 value: 81.464 - type: ndcg_at_5 value: 81.851 - type: precision_at_1 value: 67.0 - type: precision_at_10 value: 9.6 - type: precision_at_100 value: 1.0 - type: precision_at_1000 value: 0.1 - type: precision_at_20 value: 4.8500000000000005 - type: precision_at_3 value: 30.667 - type: precision_at_5 value: 18.6 - type: recall_at_1 value: 67.0 - type: recall_at_10 value: 96.0 - type: recall_at_100 value: 100.0 - type: recall_at_1000 value: 100.0 - type: recall_at_20 value: 97.0 - type: recall_at_3 value: 92.0 - type: recall_at_5 value: 93.0 - task: type: Retrieval dataset: type: jinaai/xpqa name: MTEB XPQARetrieval (fr) config: fr split: test revision: c99d599f0a6ab9b85b065da6f9d94f9cf731679f metrics: - type: map_at_1 value: 40.038000000000004 - type: map_at_10 value: 62.409000000000006 - type: map_at_100 value: 63.63999999999999 - type: map_at_1000 value: 63.693 - type: map_at_20 value: 63.165000000000006 - type: map_at_3 value: 55.364999999999995 - type: map_at_5 value: 59.95399999999999 - type: mrr_at_1 value: 62.88384512683578 - type: mrr_at_10 value: 70.414944794117 - type: mrr_at_100 value: 70.85679259651413 - type: mrr_at_1000 value: 70.8680806615119 - type: mrr_at_20 value: 70.69824986774621 - type: mrr_at_3 value: 68.04628393413438 - type: mrr_at_5 value: 69.65509568313303 - type: nauc_map_at_1000_diff1 value: 47.58306966781138 - type: nauc_map_at_1000_max value: 49.99853404950863 - type: nauc_map_at_100_diff1 value: 47.5473544905194 - type: nauc_map_at_100_max value: 49.98683021023155 - type: nauc_map_at_10_diff1 value: 47.443327641163705 - type: nauc_map_at_10_max value: 49.31862257934493 - type: nauc_map_at_1_diff1 value: 55.93203426614159 - type: nauc_map_at_1_max value: 27.467436111704224 - type: nauc_map_at_20_diff1 value: 47.454162467793985 - type: nauc_map_at_20_max value: 49.715459382963765 - type: nauc_map_at_3_diff1 value: 48.910525378486874 - type: nauc_map_at_3_max value: 42.13319318718595 - type: nauc_map_at_5_diff1 value: 48.56545403298638 - type: nauc_map_at_5_max value: 47.311811085622445 - type: nauc_mrr_at_1000_diff1 value: 56.739822956274224 - type: nauc_mrr_at_1000_max value: 58.274212468278854 - type: nauc_mrr_at_100_diff1 value: 56.7308210328899 - type: nauc_mrr_at_100_max value: 58.27250671019899 - type: nauc_mrr_at_10_diff1 value: 56.647228471816405 - type: nauc_mrr_at_10_max value: 58.210342990657495 - type: nauc_mrr_at_1_diff1 value: 58.618266167104046 - type: nauc_mrr_at_1_max value: 58.55438607166539 - type: nauc_mrr_at_20_diff1 value: 56.63534799976597 - type: nauc_mrr_at_20_max value: 58.17181317797869 - type: nauc_mrr_at_3_diff1 value: 56.815531582264825 - type: nauc_mrr_at_3_max value: 58.32821204695344 - type: nauc_mrr_at_5_diff1 value: 56.79122022985127 - type: nauc_mrr_at_5_max value: 58.20366609452701 - type: nauc_ndcg_at_1000_diff1 value: 49.530062263932194 - type: nauc_ndcg_at_1000_max value: 53.473298956705925 - type: nauc_ndcg_at_100_diff1 value: 48.95703823297219 - type: nauc_ndcg_at_100_max value: 53.191721124797276 - type: nauc_ndcg_at_10_diff1 value: 47.98530786084638 - type: nauc_ndcg_at_10_max value: 51.155857323188016 - type: nauc_ndcg_at_1_diff1 value: 58.618266167104046 - type: nauc_ndcg_at_1_max value: 58.55438607166539 - type: nauc_ndcg_at_20_diff1 value: 47.95544792051313 - type: nauc_ndcg_at_20_max value: 51.751640167194054 - type: nauc_ndcg_at_3_diff1 value: 48.50900656884395 - type: nauc_ndcg_at_3_max value: 50.78667595293348 - type: nauc_ndcg_at_5_diff1 value: 49.496100926859654 - type: nauc_ndcg_at_5_max value: 49.089893886856835 - type: nauc_precision_at_1000_diff1 value: -19.085707327488784 - type: nauc_precision_at_1000_max value: 22.16522736611267 - type: nauc_precision_at_100_diff1 value: -16.92930793417545 - type: nauc_precision_at_100_max value: 26.119556898620655 - type: nauc_precision_at_10_diff1 value: -8.586758571265364 - type: nauc_precision_at_10_max value: 34.29909350105018 - type: nauc_precision_at_1_diff1 value: 58.618266167104046 - type: nauc_precision_at_1_max value: 58.55438607166539 - type: nauc_precision_at_20_diff1 value: -12.36545815755639 - type: nauc_precision_at_20_max value: 30.779202784243694 - type: nauc_precision_at_3_diff1 value: 7.173290556095678 - type: nauc_precision_at_3_max value: 43.244915594569356 - type: nauc_precision_at_5_diff1 value: -0.5308831428158323 - type: nauc_precision_at_5_max value: 39.78478615216909 - type: nauc_recall_at_1000_diff1 value: 44.67738158424653 - type: nauc_recall_at_1000_max value: 71.12276250795361 - type: nauc_recall_at_100_diff1 value: 30.071917991701135 - type: nauc_recall_at_100_max value: 42.226214389979326 - type: nauc_recall_at_10_diff1 value: 36.275167481806804 - type: nauc_recall_at_10_max value: 40.16796727800884 - type: nauc_recall_at_1_diff1 value: 55.93203426614159 - type: nauc_recall_at_1_max value: 27.467436111704224 - type: nauc_recall_at_20_diff1 value: 32.189427460851505 - type: nauc_recall_at_20_max value: 38.926081167758205 - type: nauc_recall_at_3_diff1 value: 43.959378195689894 - type: nauc_recall_at_3_max value: 36.441633750156335 - type: nauc_recall_at_5_diff1 value: 42.6274479464408 - type: nauc_recall_at_5_max value: 38.9902118898862 - type: ndcg_at_1 value: 62.88399999999999 - type: ndcg_at_10 value: 68.907 - type: ndcg_at_100 value: 72.896 - type: ndcg_at_1000 value: 73.721 - type: ndcg_at_20 value: 70.738 - type: ndcg_at_3 value: 62.731 - type: ndcg_at_5 value: 65.191 - type: precision_at_1 value: 62.88399999999999 - type: precision_at_10 value: 16.101 - type: precision_at_100 value: 1.951 - type: precision_at_1000 value: 0.20600000000000002 - type: precision_at_20 value: 8.705 - type: precision_at_3 value: 38.095 - type: precision_at_5 value: 27.904 - type: recall_at_1 value: 40.038000000000004 - type: recall_at_10 value: 79.237 - type: recall_at_100 value: 94.17699999999999 - type: recall_at_1000 value: 99.466 - type: recall_at_20 value: 85.027 - type: recall_at_3 value: 60.336 - type: recall_at_5 value: 70.122 license: apache-2.0 language: - fr metrics: - pearsonr - spearmanr --- # [bilingual-embedding-large](https://huggingface.co/Lajavaness/bilingual-embedding-large) bilingual-embedding is the Embedding Model for bilingual language: french and english. This model is a specialized sentence-embedding trained specifically for the bilingual language, leveraging the robust capabilities of [BGE M3](https://huggingface.co/BAAI/bge-m3), a pre-trained language model larged on the [BGE M3](https://huggingface.co/BAAI/bge-m3) architecture. The model utilizes xlm-roberta to encode english-french sentences into a 1024-dimensional vector space, facilitating a wide range of applications from semantic search to text clustering. The embeddings capture the nuanced meanings of english-french sentences, reflecting both the lexical and contextual layers of the language. ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BilingualModel (1): Pooling({'word_embedding_dimension': 1024, '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}) (2): Normalize() ) ``` ## Training and Fine-tuning process #### Stage 1: NLI Training - Dataset: [(SNLI+XNLI) for english+french] - Method: Training using Multi-Negative Ranking Loss. This stage focused on improving the model's ability to discern and rank nuanced differences in sentence semantics. ### Stage 3: Continued Fine-tuning for Semantic Textual Similarity on STS Benchmark - Dataset: [STSB-fr and en] - Method: Fine-tuning specifically for the semantic textual similarity benchmark using Siamese BERT-Networks configured with the 'sentence-transformers' library. ### Stage 4: Advanced Augmentation Fine-tuning - Dataset: STSB with generate [silver sample from gold sample](https://www.sbert.net/examples/training/data_augmentation/README.html) - Method: Employed an advanced strategy using [Augmented SBERT](https://arxiv.org/abs/2010.08240) with Pair Sampling Strategies, integrating both Cross-Encoder and Bi-Encoder models. This stage further refined the embeddings by enriching the training data dynamically, enhancing the model's robustness and accuracy. ## Usage: Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["Paris est une capitale de la France", "Paris is a capital of France"] model = SentenceTransformer('Lajavaness/bilingual-embedding-large-8k', trust_remote_code=True) print(embeddings) ``` ## Evaluation TODO ## Citation @article{chen2024bge, title={Bge m3-embedding: Multi-lingual, multi-functionality, multi-granularity text embeddings through self-knowledge distillation}, author={Chen, Jianlv and Xiao, Shitao and Zhang, Peitian and Luo, Kun and Lian, Defu and Liu, Zheng}, journal={arXiv preprint arXiv:2402.03216}, year={2024} } @article{conneau2019unsupervised, title={Unsupervised cross-lingual representation learning at scale}, author={Conneau, Alexis and Khandelwal, Kartikay and Goyal, Naman and Chaudhary, Vishrav and Wenzek, Guillaume and Guzm{\'a}n, Francisco and Grave, Edouard and Ott, Myle and Zettlemoyer, Luke and Stoyanov, Veselin}, journal={arXiv preprint arXiv:1911.02116}, year={2019} } @article{reimers2019sentence, title={Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks}, author={Nils Reimers, Iryna Gurevych}, journal={https://arxiv.org/abs/1908.10084}, year={2019} } @article{thakur2020augmented, title={Augmented SBERT: Data Augmentation Method for Improving Bi-Encoders for Pairwise Sentence Scoring Tasks}, author={Thakur, Nandan and Reimers, Nils and Daxenberger, Johannes and Gurevych, Iryna}, journal={arXiv e-prints}, pages={arXiv--2010}, year={2020}