tuan.ljn commited on
Commit
23ff5a5
·
1 Parent(s): ee0e482

Add: update model

Browse files
.ipynb_checkpoints/README-checkpoint.md DELETED
@@ -1,1558 +0,0 @@
1
- ---
2
- library_name: sentence-transformers
3
- pipeline_tag: sentence-similarity
4
- tags:
5
- - sentence-transformers
6
- - feature-extraction
7
- - sentence-similarity
8
- - transformers
9
- - sentence-embedding
10
- - mteb
11
- model-index:
12
- - name: fea4158624861f232b237be2e106dc7d8295e0da
13
- results:
14
- - task:
15
- type: Clustering
16
- dataset:
17
- type: lyon-nlp/alloprof
18
- name: MTEB AlloProfClusteringP2P
19
- config: default
20
- split: test
21
- revision: 392ba3f5bcc8c51f578786c1fc3dae648662cb9b
22
- metrics:
23
- - type: v_measure
24
- value: 56.77190187231352
25
- - type: v_measures
26
- value: [0.5591529760439443, 0.5974955147482336, 0.5823433887463434, 0.5582030533486634, 0.5608784889863708]
27
- - task:
28
- type: Clustering
29
- dataset:
30
- type: lyon-nlp/alloprof
31
- name: MTEB AlloProfClusteringS2S
32
- config: default
33
- split: test
34
- revision: 392ba3f5bcc8c51f578786c1fc3dae648662cb9b
35
- metrics:
36
- - type: v_measure
37
- value: 43.8035296814567
38
- - type: v_measures
39
- value: [0.48563274496248593, 0.39879051227524587, 0.4846924607458884, 0.48918774336682036, 0.42626900366019754]
40
- - task:
41
- type: Reranking
42
- dataset:
43
- type: lyon-nlp/mteb-fr-reranking-alloprof-s2p
44
- name: MTEB AlloprofReranking
45
- config: default
46
- split: test
47
- revision: 65393d0d7a08a10b4e348135e824f385d420b0fd
48
- metrics:
49
- - type: map
50
- value: 71.51753654257872
51
- - type: mrr
52
- value: 72.59564245963209
53
- - type: nAUC_map_diff1
54
- value: 54.75803232873865
55
- - type: nAUC_map_max
56
- value: 22.763024994564173
57
- - type: nAUC_mrr_diff1
58
- value: 54.34234068011684
59
- - type: nAUC_mrr_max
60
- value: 23.607169979872587
61
- - task:
62
- type: Retrieval
63
- dataset:
64
- type: lyon-nlp/alloprof
65
- name: MTEB AlloprofRetrieval
66
- config: default
67
- split: test
68
- revision: fcf295ea64c750f41fadbaa37b9b861558e1bfbd
69
- metrics:
70
- - type: map_at_1
71
- value: 26.598
72
- - type: map_at_10
73
- value: 37.808
74
- - type: map_at_100
75
- value: 38.726
76
- - type: map_at_1000
77
- value: 38.778
78
- - type: map_at_20
79
- value: 38.34
80
- - type: map_at_3
81
- value: 34.599999999999994
82
- - type: map_at_5
83
- value: 36.385
84
- - type: mrr_at_1
85
- value: 26.59758203799655
86
- - type: mrr_at_10
87
- value: 37.80750541437081
88
- - type: mrr_at_100
89
- value: 38.72559086585011
90
- - type: mrr_at_1000
91
- value: 38.77782362787768
92
- - type: mrr_at_20
93
- value: 38.33973921272315
94
- - type: mrr_at_3
95
- value: 34.59988485895229
96
- - type: mrr_at_5
97
- value: 36.385290731145794
98
- - type: nauc_map_at_1000_diff1
99
- value: 35.14662281558396
100
- - type: nauc_map_at_1000_max
101
- value: 33.29444147034975
102
- - type: nauc_map_at_100_diff1
103
- value: 35.14901176407076
104
- - type: nauc_map_at_100_max
105
- value: 33.32860103491456
106
- - type: nauc_map_at_10_diff1
107
- value: 34.96490139273049
108
- - type: nauc_map_at_10_max
109
- value: 33.252929811567526
110
- - type: nauc_map_at_1_diff1
111
- value: 40.652132824871664
112
- - type: nauc_map_at_1_max
113
- value: 28.53606743237387
114
- - type: nauc_map_at_20_diff1
115
- value: 35.11177701050558
116
- - type: nauc_map_at_20_max
117
- value: 33.38822815064973
118
- - type: nauc_map_at_3_diff1
119
- value: 35.08068505968589
120
- - type: nauc_map_at_3_max
121
- value: 32.10125944853496
122
- - type: nauc_map_at_5_diff1
123
- value: 34.78702330258393
124
- - type: nauc_map_at_5_max
125
- value: 32.89738895858572
126
- - type: nauc_mrr_at_1000_diff1
127
- value: 35.14662281558396
128
- - type: nauc_mrr_at_1000_max
129
- value: 33.29444147034975
130
- - type: nauc_mrr_at_100_diff1
131
- value: 35.14901176407076
132
- - type: nauc_mrr_at_100_max
133
- value: 33.32860103491456
134
- - type: nauc_mrr_at_10_diff1
135
- value: 34.96490139273049
136
- - type: nauc_mrr_at_10_max
137
- value: 33.252929811567526
138
- - type: nauc_mrr_at_1_diff1
139
- value: 40.652132824871664
140
- - type: nauc_mrr_at_1_max
141
- value: 28.53606743237387
142
- - type: nauc_mrr_at_20_diff1
143
- value: 35.11177701050558
144
- - type: nauc_mrr_at_20_max
145
- value: 33.38822815064973
146
- - type: nauc_mrr_at_3_diff1
147
- value: 35.08068505968589
148
- - type: nauc_mrr_at_3_max
149
- value: 32.10125944853496
150
- - type: nauc_mrr_at_5_diff1
151
- value: 34.78702330258393
152
- - type: nauc_mrr_at_5_max
153
- value: 32.89738895858572
154
- - type: nauc_ndcg_at_1000_diff1
155
- value: 34.24563790011671
156
- - type: nauc_ndcg_at_1000_max
157
- value: 35.1750183970367
158
- - type: nauc_ndcg_at_100_diff1
159
- value: 34.26540576121903
160
- - type: nauc_ndcg_at_100_max
161
- value: 36.211723696019526
162
- - type: nauc_ndcg_at_10_diff1
163
- value: 33.363793757214985
164
- - type: nauc_ndcg_at_10_max
165
- value: 35.724249117130285
166
- - type: nauc_ndcg_at_1_diff1
167
- value: 40.652132824871664
168
- - type: nauc_ndcg_at_1_max
169
- value: 28.53606743237387
170
- - type: nauc_ndcg_at_20_diff1
171
- value: 33.82271561239704
172
- - type: nauc_ndcg_at_20_max
173
- value: 36.31393467921569
174
- - type: nauc_ndcg_at_3_diff1
175
- value: 33.47602274641705
176
- - type: nauc_ndcg_at_3_max
177
- value: 33.25855418368982
178
- - type: nauc_ndcg_at_5_diff1
179
- value: 32.982858998932784
180
- - type: nauc_ndcg_at_5_max
181
- value: 34.683197330270694
182
- - type: nauc_precision_at_1000_diff1
183
- value: 20.695105766988465
184
- - type: nauc_precision_at_1000_max
185
- value: 79.24780459990231
186
- - type: nauc_precision_at_100_diff1
187
- value: 32.1302666444128
188
- - type: nauc_precision_at_100_max
189
- value: 59.829884850113594
190
- - type: nauc_precision_at_10_diff1
191
- value: 28.35768310864352
192
- - type: nauc_precision_at_10_max
193
- value: 44.56445585989902
194
- - type: nauc_precision_at_1_diff1
195
- value: 40.652132824871664
196
- - type: nauc_precision_at_1_max
197
- value: 28.53606743237387
198
- - type: nauc_precision_at_20_diff1
199
- value: 29.66063385017264
200
- - type: nauc_precision_at_20_max
201
- value: 48.87458861994212
202
- - type: nauc_precision_at_3_diff1
203
- value: 29.054817586181176
204
- - type: nauc_precision_at_3_max
205
- value: 36.488441946705876
206
- - type: nauc_precision_at_5_diff1
207
- value: 27.79863250059474
208
- - type: nauc_precision_at_5_max
209
- value: 40.02591480379938
210
- - type: nauc_recall_at_1000_diff1
211
- value: 20.695105766987187
212
- - type: nauc_recall_at_1000_max
213
- value: 79.24780459990147
214
- - type: nauc_recall_at_100_diff1
215
- value: 32.13026664441275
216
- - type: nauc_recall_at_100_max
217
- value: 59.829884850113736
218
- - type: nauc_recall_at_10_diff1
219
- value: 28.357683108643496
220
- - type: nauc_recall_at_10_max
221
- value: 44.56445585989908
222
- - type: nauc_recall_at_1_diff1
223
- value: 40.652132824871664
224
- - type: nauc_recall_at_1_max
225
- value: 28.53606743237387
226
- - type: nauc_recall_at_20_diff1
227
- value: 29.660633850172687
228
- - type: nauc_recall_at_20_max
229
- value: 48.87458861994213
230
- - type: nauc_recall_at_3_diff1
231
- value: 29.05481758618121
232
- - type: nauc_recall_at_3_max
233
- value: 36.48844194670593
234
- - type: nauc_recall_at_5_diff1
235
- value: 27.798632500594728
236
- - type: nauc_recall_at_5_max
237
- value: 40.025914803799395
238
- - type: ndcg_at_1
239
- value: 26.598
240
- - type: ndcg_at_10
241
- value: 43.902
242
- - type: ndcg_at_100
243
- value: 48.647
244
- - type: ndcg_at_1000
245
- value: 50.135
246
- - type: ndcg_at_20
247
- value: 45.794000000000004
248
- - type: ndcg_at_3
249
- value: 37.233
250
- - type: ndcg_at_5
251
- value: 40.47
252
- - type: precision_at_1
253
- value: 26.598
254
- - type: precision_at_10
255
- value: 6.334
256
- - type: precision_at_100
257
- value: 0.8619999999999999
258
- - type: precision_at_1000
259
- value: 0.098
260
- - type: precision_at_20
261
- value: 3.5360000000000005
262
- - type: precision_at_3
263
- value: 14.954
264
- - type: precision_at_5
265
- value: 10.552999999999999
266
- - type: recall_at_1
267
- value: 26.598
268
- - type: recall_at_10
269
- value: 63.342
270
- - type: recall_at_100
271
- value: 86.226
272
- - type: recall_at_1000
273
- value: 98.143
274
- - type: recall_at_20
275
- value: 70.72500000000001
276
- - type: recall_at_3
277
- value: 44.862
278
- - type: recall_at_5
279
- value: 52.763000000000005
280
- - task:
281
- type: Classification
282
- dataset:
283
- type: mteb/amazon_reviews_multi
284
- name: MTEB AmazonReviewsClassification
285
- config: default
286
- split: test
287
- revision: 1399c76144fd37290681b995c656ef9b2e06e26d
288
- metrics:
289
- - type: accuracy
290
- value: 45.504
291
- - type: f1
292
- value: 43.653559521530944
293
- - type: f1_weighted
294
- value: 43.65355952153093
295
- - task:
296
- type: Retrieval
297
- dataset:
298
- type: maastrichtlawtech/bsard
299
- name: MTEB BSARDRetrieval
300
- config: default
301
- split: test
302
- revision: 5effa1b9b5fa3b0f9e12523e6e43e5f86a6e6d59
303
- metrics:
304
- - type: map_at_1
305
- value: 6.757000000000001
306
- - type: map_at_10
307
- value: 10.569
308
- - type: map_at_100
309
- value: 11.641
310
- - type: map_at_1000
311
- value: 11.758000000000001
312
- - type: map_at_20
313
- value: 11.152
314
- - type: map_at_3
315
- value: 9.009
316
- - type: map_at_5
317
- value: 9.82
318
- - type: mrr_at_1
319
- value: 6.756756756756757
320
- - type: mrr_at_10
321
- value: 10.568604318604317
322
- - type: mrr_at_100
323
- value: 11.640572409499667
324
- - type: mrr_at_1000
325
- value: 11.757590970218725
326
- - type: mrr_at_20
327
- value: 11.152253605813977
328
- - type: mrr_at_3
329
- value: 9.00900900900901
330
- - type: mrr_at_5
331
- value: 9.819819819819822
332
- - type: nauc_map_at_1000_diff1
333
- value: 20.497861726027473
334
- - type: nauc_map_at_1000_max
335
- value: 8.515289767591149
336
- - type: nauc_map_at_100_diff1
337
- value: 20.457264311409304
338
- - type: nauc_map_at_100_max
339
- value: 8.478315371126714
340
- - type: nauc_map_at_10_diff1
341
- value: 22.122755317386826
342
- - type: nauc_map_at_10_max
343
- value: 8.684832816651243
344
- - type: nauc_map_at_1_diff1
345
- value: 32.7324170733489
346
- - type: nauc_map_at_1_max
347
- value: 17.632507133954086
348
- - type: nauc_map_at_20_diff1
349
- value: 21.309144510706552
350
- - type: nauc_map_at_20_max
351
- value: 8.357569194331324
352
- - type: nauc_map_at_3_diff1
353
- value: 24.640888091380244
354
- - type: nauc_map_at_3_max
355
- value: 8.512417316260153
356
- - type: nauc_map_at_5_diff1
357
- value: 23.696706451320555
358
- - type: nauc_map_at_5_max
359
- value: 10.065581499162409
360
- - type: nauc_mrr_at_1000_diff1
361
- value: 20.497861726027473
362
- - type: nauc_mrr_at_1000_max
363
- value: 8.515289767591149
364
- - type: nauc_mrr_at_100_diff1
365
- value: 20.457264311409304
366
- - type: nauc_mrr_at_100_max
367
- value: 8.478315371126714
368
- - type: nauc_mrr_at_10_diff1
369
- value: 22.122755317386826
370
- - type: nauc_mrr_at_10_max
371
- value: 8.684832816651243
372
- - type: nauc_mrr_at_1_diff1
373
- value: 32.7324170733489
374
- - type: nauc_mrr_at_1_max
375
- value: 17.632507133954086
376
- - type: nauc_mrr_at_20_diff1
377
- value: 21.309144510706552
378
- - type: nauc_mrr_at_20_max
379
- value: 8.357569194331324
380
- - type: nauc_mrr_at_3_diff1
381
- value: 24.640888091380244
382
- - type: nauc_mrr_at_3_max
383
- value: 8.512417316260153
384
- - type: nauc_mrr_at_5_diff1
385
- value: 23.696706451320555
386
- - type: nauc_mrr_at_5_max
387
- value: 10.065581499162409
388
- - type: nauc_ndcg_at_1000_diff1
389
- value: 13.492135744038377
390
- - type: nauc_ndcg_at_1000_max
391
- value: 9.019754831261519
392
- - type: nauc_ndcg_at_100_diff1
393
- value: 12.386959698428296
394
- - type: nauc_ndcg_at_100_max
395
- value: 8.140082932773288
396
- - type: nauc_ndcg_at_10_diff1
397
- value: 18.08185602779908
398
- - type: nauc_ndcg_at_10_max
399
- value: 6.451070792965509
400
- - type: nauc_ndcg_at_1_diff1
401
- value: 32.7324170733489
402
- - type: nauc_ndcg_at_1_max
403
- value: 17.632507133954086
404
- - type: nauc_ndcg_at_20_diff1
405
- value: 16.104176022358285
406
- - type: nauc_ndcg_at_20_max
407
- value: 5.670070730016123
408
- - type: nauc_ndcg_at_3_diff1
409
- value: 22.51956353681352
410
- - type: nauc_ndcg_at_3_max
411
- value: 6.154988622749747
412
- - type: nauc_ndcg_at_5_diff1
413
- value: 20.90624606199523
414
- - type: nauc_ndcg_at_5_max
415
- value: 9.018795972091642
416
- - type: nauc_precision_at_1000_diff1
417
- value: -7.5156227166904666
418
- - type: nauc_precision_at_1000_max
419
- value: 21.389191162087265
420
- - type: nauc_precision_at_100_diff1
421
- value: -3.485942444137334
422
- - type: nauc_precision_at_100_max
423
- value: 11.604808001151612
424
- - type: nauc_precision_at_10_diff1
425
- value: 10.569691962167767
426
- - type: nauc_precision_at_10_max
427
- value: 2.5249979762023176
428
- - type: nauc_precision_at_1_diff1
429
- value: 32.7324170733489
430
- - type: nauc_precision_at_1_max
431
- value: 17.632507133954086
432
- - type: nauc_precision_at_20_diff1
433
- value: 6.993263678069275
434
- - type: nauc_precision_at_20_max
435
- value: 1.212342757686577
436
- - type: nauc_precision_at_3_diff1
437
- value: 17.747702545254146
438
- - type: nauc_precision_at_3_max
439
- value: 0.8595742156164364
440
- - type: nauc_precision_at_5_diff1
441
- value: 15.109328976502349
442
- - type: nauc_precision_at_5_max
443
- value: 7.278141846526832
444
- - type: nauc_recall_at_1000_diff1
445
- value: -7.515622716690405
446
- - type: nauc_recall_at_1000_max
447
- value: 21.389191162087183
448
- - type: nauc_recall_at_100_diff1
449
- value: -3.4859424441373585
450
- - type: nauc_recall_at_100_max
451
- value: 11.604808001151582
452
- - type: nauc_recall_at_10_diff1
453
- value: 10.569691962167727
454
- - type: nauc_recall_at_10_max
455
- value: 2.524997976202258
456
- - type: nauc_recall_at_1_diff1
457
- value: 32.7324170733489
458
- - type: nauc_recall_at_1_max
459
- value: 17.632507133954086
460
- - type: nauc_recall_at_20_diff1
461
- value: 6.9932636780692325
462
- - type: nauc_recall_at_20_max
463
- value: 1.2123427576865247
464
- - type: nauc_recall_at_3_diff1
465
- value: 17.74770254525415
466
- - type: nauc_recall_at_3_max
467
- value: 0.8595742156164401
468
- - type: nauc_recall_at_5_diff1
469
- value: 15.109328976502375
470
- - type: nauc_recall_at_5_max
471
- value: 7.278141846526856
472
- - type: ndcg_at_1
473
- value: 6.757000000000001
474
- - type: ndcg_at_10
475
- value: 13.147
476
- - type: ndcg_at_100
477
- value: 18.932
478
- - type: ndcg_at_1000
479
- value: 22.663
480
- - type: ndcg_at_20
481
- value: 15.222
482
- - type: ndcg_at_3
483
- value: 9.812999999999999
484
- - type: ndcg_at_5
485
- value: 11.286
486
- - type: precision_at_1
487
- value: 6.757000000000001
488
- - type: precision_at_10
489
- value: 2.162
490
- - type: precision_at_100
491
- value: 0.5
492
- - type: precision_at_1000
493
- value: 0.08099999999999999
494
- - type: precision_at_20
495
- value: 1.486
496
- - type: precision_at_3
497
- value: 4.054
498
- - type: precision_at_5
499
- value: 3.1530000000000005
500
- - type: recall_at_1
501
- value: 6.757000000000001
502
- - type: recall_at_10
503
- value: 21.622
504
- - type: recall_at_100
505
- value: 50.0
506
- - type: recall_at_1000
507
- value: 80.631
508
- - type: recall_at_20
509
- value: 29.73
510
- - type: recall_at_3
511
- value: 12.162
512
- - type: recall_at_5
513
- value: 15.766
514
- - task:
515
- type: Clustering
516
- dataset:
517
- type: lyon-nlp/clustering-hal-s2s
518
- name: MTEB HALClusteringS2S
519
- config: default
520
- split: test
521
- revision: e06ebbbb123f8144bef1a5d18796f3dec9ae2915
522
- metrics:
523
- - type: v_measure
524
- value: 24.858750161104126
525
- - type: v_measures
526
- value: [0.2984717048033491, 0.2554963718476627, 0.27306743677619566, 0.2655317293951285, 0.22365142310852648]
527
- - task:
528
- type: Clustering
529
- dataset:
530
- type: reciTAL/mlsum
531
- name: MTEB MLSUMClusteringP2P
532
- config: default
533
- split: test
534
- revision: b5d54f8f3b61ae17845046286940f03c6bc79bc7
535
- metrics:
536
- - type: v_measure
537
- value: 39.196489283513955
538
- - type: v_measures
539
- value: [0.397079645426394, 0.41814959265244056, 0.4075805000522318, 0.3777207448521023, 0.3499037023664506]
540
- - task:
541
- type: Clustering
542
- dataset:
543
- type: reciTAL/mlsum
544
- name: MTEB MLSUMClusteringS2S
545
- config: default
546
- split: test
547
- revision: b5d54f8f3b61ae17845046286940f03c6bc79bc7
548
- metrics:
549
- - type: v_measure
550
- value: 38.90242301200363
551
- - type: v_measures
552
- value: [0.3879152638224075, 0.4155376524170288, 0.41644489647223826, 0.3720493791140543, 0.33885028730003314]
553
- - task:
554
- type: Classification
555
- dataset:
556
- type: mteb/mtop_domain
557
- name: MTEB MTOPDomainClassification
558
- config: default
559
- split: test
560
- revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf
561
- metrics:
562
- - type: accuracy
563
- value: 86.88380833072345
564
- - type: f1
565
- value: 86.96787213648228
566
- - type: f1_weighted
567
- value: 86.83432508604707
568
- - task:
569
- type: Classification
570
- dataset:
571
- type: mteb/mtop_intent
572
- name: MTEB MTOPIntentClassification
573
- config: default
574
- split: test
575
- revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba
576
- metrics:
577
- - type: accuracy
578
- value: 63.780144065142494
579
- - type: f1
580
- value: 46.070400220183394
581
- - type: f1_weighted
582
- value: 66.2871618019472
583
- - task:
584
- type: Classification
585
- dataset:
586
- type: mteb/masakhanews
587
- name: MTEB MasakhaNEWSClassification
588
- config: default
589
- split: test
590
- revision: 18193f187b92da67168c655c9973a165ed9593dd
591
- metrics:
592
- - type: accuracy
593
- value: 73.64928909952606
594
- - type: f1
595
- value: 70.38554255346646
596
- - type: f1_weighted
597
- value: 73.78534895009892
598
- - task:
599
- type: Clustering
600
- dataset:
601
- type: masakhane/masakhanews
602
- name: MTEB MasakhaNEWSClusteringP2P
603
- config: default
604
- split: test
605
- revision: 8ccc72e69e65f40c70e117d8b3c08306bb788b60
606
- metrics:
607
- - type: v_measure
608
- value: 53.94667838037061
609
- - type: v_measures
610
- value: [1.0, 0.02235188516574002, 0.3929431892497775, 0.5064698281428253, 0.7755690164601873]
611
- - task:
612
- type: Clustering
613
- dataset:
614
- type: masakhane/masakhanews
615
- name: MTEB MasakhaNEWSClusteringS2S
616
- config: default
617
- split: test
618
- revision: 8ccc72e69e65f40c70e117d8b3c08306bb788b60
619
- metrics:
620
- - type: v_measure
621
- value: 34.0804579102654
622
- - type: v_measures
623
- value: [1.0, 0.03741154081338069, 0.3515971997960832, 0.027612964551341854, 0.2874011903524643]
624
- - task:
625
- type: Classification
626
- dataset:
627
- type: mteb/amazon_massive_intent
628
- name: MTEB MassiveIntentClassification
629
- config: default
630
- split: test
631
- revision: 4672e20407010da34463acc759c162ca9734bca6
632
- metrics:
633
- - type: accuracy
634
- value: 66.42905178211164
635
- - type: f1
636
- value: 64.35827544332014
637
- - type: f1_weighted
638
- value: 65.57615486214955
639
- - task:
640
- type: Classification
641
- dataset:
642
- type: mteb/amazon_massive_scenario
643
- name: MTEB MassiveScenarioClassification
644
- config: default
645
- split: test
646
- revision: fad2c6e8459f9e1c45d9315f4953d921437d70f8
647
- metrics:
648
- - type: accuracy
649
- value: 70.72293207800942
650
- - type: f1
651
- value: 70.02549388005589
652
- - type: f1_weighted
653
- value: 70.47189927452128
654
- - task:
655
- type: Retrieval
656
- dataset:
657
- type: jinaai/mintakaqa
658
- name: MTEB MintakaRetrieval
659
- config: default
660
- split: test
661
- revision: efa78cc2f74bbcd21eff2261f9e13aebe40b814e
662
- metrics:
663
- - type: map_at_1
664
- value: 19.41
665
- - type: map_at_10
666
- value: 28.254
667
- - type: map_at_100
668
- value: 29.355999999999998
669
- - type: map_at_1000
670
- value: 29.444
671
- - type: map_at_20
672
- value: 28.918
673
- - type: map_at_3
674
- value: 25.833000000000002
675
- - type: map_at_5
676
- value: 27.141
677
- - type: mrr_at_1
678
- value: 19.41031941031941
679
- - type: mrr_at_10
680
- value: 28.254309504309553
681
- - type: mrr_at_100
682
- value: 29.35551984998816
683
- - type: mrr_at_1000
684
- value: 29.44409433115412
685
- - type: mrr_at_20
686
- value: 28.91784141291608
687
- - type: mrr_at_3
688
- value: 25.832650832650856
689
- - type: mrr_at_5
690
- value: 27.141004641004695
691
- - type: nauc_map_at_1000_diff1
692
- value: 21.162510480664746
693
- - type: nauc_map_at_1000_max
694
- value: 30.110199608609662
695
- - type: nauc_map_at_100_diff1
696
- value: 21.14297046777571
697
- - type: nauc_map_at_100_max
698
- value: 30.122604363568612
699
- - type: nauc_map_at_10_diff1
700
- value: 21.199056431758258
701
- - type: nauc_map_at_10_max
702
- value: 30.270246884757647
703
- - type: nauc_map_at_1_diff1
704
- value: 26.23217963475422
705
- - type: nauc_map_at_1_max
706
- value: 25.259209130612533
707
- - type: nauc_map_at_20_diff1
708
- value: 21.13743075881012
709
- - type: nauc_map_at_20_max
710
- value: 30.22788200140141
711
- - type: nauc_map_at_3_diff1
712
- value: 21.972972142629406
713
- - type: nauc_map_at_3_max
714
- value: 29.75552354821982
715
- - type: nauc_map_at_5_diff1
716
- value: 21.79084317530715
717
- - type: nauc_map_at_5_max
718
- value: 30.387209515342473
719
- - type: nauc_mrr_at_1000_diff1
720
- value: 21.162510480664746
721
- - type: nauc_mrr_at_1000_max
722
- value: 30.110199608609662
723
- - type: nauc_mrr_at_100_diff1
724
- value: 21.14297046777571
725
- - type: nauc_mrr_at_100_max
726
- value: 30.122604363568612
727
- - type: nauc_mrr_at_10_diff1
728
- value: 21.199056431758258
729
- - type: nauc_mrr_at_10_max
730
- value: 30.270246884757647
731
- - type: nauc_mrr_at_1_diff1
732
- value: 26.23217963475422
733
- - type: nauc_mrr_at_1_max
734
- value: 25.259209130612533
735
- - type: nauc_mrr_at_20_diff1
736
- value: 21.13743075881012
737
- - type: nauc_mrr_at_20_max
738
- value: 30.22788200140141
739
- - type: nauc_mrr_at_3_diff1
740
- value: 21.972972142629406
741
- - type: nauc_mrr_at_3_max
742
- value: 29.75552354821982
743
- - type: nauc_mrr_at_5_diff1
744
- value: 21.79084317530715
745
- - type: nauc_mrr_at_5_max
746
- value: 30.387209515342473
747
- - type: nauc_ndcg_at_1000_diff1
748
- value: 19.513865750647934
749
- - type: nauc_ndcg_at_1000_max
750
- value: 30.88832578481811
751
- - type: nauc_ndcg_at_100_diff1
752
- value: 18.80377362018204
753
- - type: nauc_ndcg_at_100_max
754
- value: 30.952837388928288
755
- - type: nauc_ndcg_at_10_diff1
756
- value: 19.020409577228836
757
- - type: nauc_ndcg_at_10_max
758
- value: 31.70346401198393
759
- - type: nauc_ndcg_at_1_diff1
760
- value: 26.23217963475422
761
- - type: nauc_ndcg_at_1_max
762
- value: 25.259209130612533
763
- - type: nauc_ndcg_at_20_diff1
764
- value: 18.76950183960116
765
- - type: nauc_ndcg_at_20_max
766
- value: 31.598953492190745
767
- - type: nauc_ndcg_at_3_diff1
768
- value: 20.824179655562357
769
- - type: nauc_ndcg_at_3_max
770
- value: 31.0541305570042
771
- - type: nauc_ndcg_at_5_diff1
772
- value: 20.483978673834002
773
- - type: nauc_ndcg_at_5_max
774
- value: 32.10721692420019
775
- - type: nauc_precision_at_1000_diff1
776
- value: 2.2196094973599374
777
- - type: nauc_precision_at_1000_max
778
- value: 40.25789000420308
779
- - type: nauc_precision_at_100_diff1
780
- value: 8.63682048375218
781
- - type: nauc_precision_at_100_max
782
- value: 32.32052516290328
783
- - type: nauc_precision_at_10_diff1
784
- value: 12.717058324435426
785
- - type: nauc_precision_at_10_max
786
- value: 35.265251512978985
787
- - type: nauc_precision_at_1_diff1
788
- value: 26.23217963475422
789
- - type: nauc_precision_at_1_max
790
- value: 25.259209130612533
791
- - type: nauc_precision_at_20_diff1
792
- value: 11.220279583941753
793
- - type: nauc_precision_at_20_max
794
- value: 35.20392547315769
795
- - type: nauc_precision_at_3_diff1
796
- value: 17.863403819374867
797
- - type: nauc_precision_at_3_max
798
- value: 34.42645261456197
799
- - type: nauc_precision_at_5_diff1
800
- value: 17.114444816553625
801
- - type: nauc_precision_at_5_max
802
- value: 36.59616112935629
803
- - type: nauc_recall_at_1000_diff1
804
- value: 2.219609497359936
805
- - type: nauc_recall_at_1000_max
806
- value: 40.257890004202366
807
- - type: nauc_recall_at_100_diff1
808
- value: 8.63682048375225
809
- - type: nauc_recall_at_100_max
810
- value: 32.320525162903365
811
- - type: nauc_recall_at_10_diff1
812
- value: 12.71705832443547
813
- - type: nauc_recall_at_10_max
814
- value: 35.26525151297903
815
- - type: nauc_recall_at_1_diff1
816
- value: 26.23217963475422
817
- - type: nauc_recall_at_1_max
818
- value: 25.259209130612533
819
- - type: nauc_recall_at_20_diff1
820
- value: 11.220279583941677
821
- - type: nauc_recall_at_20_max
822
- value: 35.203925473157646
823
- - type: nauc_recall_at_3_diff1
824
- value: 17.86340381937489
825
- - type: nauc_recall_at_3_max
826
- value: 34.42645261456199
827
- - type: nauc_recall_at_5_diff1
828
- value: 17.11444481655362
829
- - type: nauc_recall_at_5_max
830
- value: 36.59616112935631
831
- - type: ndcg_at_1
832
- value: 19.41
833
- - type: ndcg_at_10
834
- value: 32.956
835
- - type: ndcg_at_100
836
- value: 38.504
837
- - type: ndcg_at_1000
838
- value: 41.217
839
- - type: ndcg_at_20
840
- value: 35.35
841
- - type: ndcg_at_3
842
- value: 27.898
843
- - type: ndcg_at_5
844
- value: 30.249
845
- - type: precision_at_1
846
- value: 19.41
847
- - type: precision_at_10
848
- value: 4.795
849
- - type: precision_at_100
850
- value: 0.744
851
- - type: precision_at_1000
852
- value: 0.096
853
- - type: precision_at_20
854
- value: 2.869
855
- - type: precision_at_3
856
- value: 11.289
857
- - type: precision_at_5
858
- value: 7.912
859
- - type: recall_at_1
860
- value: 19.41
861
- - type: recall_at_10
862
- value: 47.952
863
- - type: recall_at_100
864
- value: 74.36500000000001
865
- - type: recall_at_1000
866
- value: 96.478
867
- - type: recall_at_20
868
- value: 57.371
869
- - type: recall_at_3
870
- value: 33.866
871
- - type: recall_at_5
872
- value: 39.558
873
- - task:
874
- type: PairClassification
875
- dataset:
876
- type: GEM/opusparcus
877
- name: MTEB OpusparcusPC
878
- config: default
879
- split: test
880
- revision: 9e9b1f8ef51616073f47f306f7f47dd91663f86a
881
- metrics: []
882
- - task:
883
- type: PairClassification
884
- dataset:
885
- type: google-research-datasets/paws-x
886
- name: MTEB PawsX
887
- config: default
888
- split: test
889
- revision: 8a04d940a42cd40658986fdd8e3da561533a3646
890
- metrics:
891
- - type: cos_sim_accuracy
892
- value: 64.25
893
- - type: cos_sim_ap
894
- value: 64.3383111759356
895
- - type: cos_sim_f1
896
- value: 63.03703703703704
897
- - type: cos_sim_precision
898
- value: 47.35670562047857
899
- - type: cos_sim_recall
900
- value: 94.24141749723145
901
- - type: dot_accuracy
902
- value: 64.25
903
- - type: dot_ap
904
- value: 64.33647413039195
905
- - type: dot_f1
906
- value: 63.03703703703704
907
- - type: dot_precision
908
- value: 47.35670562047857
909
- - type: dot_recall
910
- value: 94.24141749723145
911
- - type: euclidean_accuracy
912
- value: 64.25
913
- - type: euclidean_ap
914
- value: 64.33837256418407
915
- - type: euclidean_f1
916
- value: 63.03703703703704
917
- - type: euclidean_precision
918
- value: 47.35670562047857
919
- - type: euclidean_recall
920
- value: 94.24141749723145
921
- - type: manhattan_accuracy
922
- value: 64.3
923
- - type: manhattan_ap
924
- value: 64.32795078191543
925
- - type: manhattan_f1
926
- value: 62.935507783543365
927
- - type: manhattan_precision
928
- value: 47.298050139275766
929
- - type: manhattan_recall
930
- value: 94.01993355481729
931
- - type: max_accuracy
932
- value: 64.3
933
- - type: max_ap
934
- value: 64.33837256418407
935
- - type: max_f1
936
- value: 63.03703703703704
937
- - task:
938
- type: STS
939
- dataset:
940
- type: Lajavaness/SICK-fr
941
- name: MTEB SICKFr
942
- config: default
943
- split: test
944
- revision: e077ab4cf4774a1e36d86d593b150422fafd8e8a
945
- metrics:
946
- - type: cos_sim_pearson
947
- value: 84.72950355896789
948
- - type: cos_sim_spearman
949
- value: 79.4608290812696
950
- - type: euclidean_pearson
951
- value: 81.32539142627735
952
- - type: euclidean_spearman
953
- value: 79.46019403205177
954
- - type: manhattan_pearson
955
- value: 81.35998006674681
956
- - type: manhattan_spearman
957
- value: 79.41438354042496
958
- - task:
959
- type: STS
960
- dataset:
961
- type: mteb/sts22-crosslingual-sts
962
- name: MTEB STS22
963
- config: default
964
- split: test
965
- revision: de9d86b3b84231dc21f76c7b7af1f28e2f57f6e3
966
- metrics:
967
- - type: cos_sim_pearson
968
- value: 81.94721498063055
969
- - type: cos_sim_spearman
970
- value: 83.27561639335909
971
- - type: euclidean_pearson
972
- value: 80.74250472409508
973
- - type: euclidean_spearman
974
- value: 83.27561639335909
975
- - type: manhattan_pearson
976
- value: 80.55336080634422
977
- - type: manhattan_spearman
978
- value: 83.46556509775091
979
- - task:
980
- type: STS
981
- dataset:
982
- type: mteb/stsb_multi_mt
983
- name: MTEB STSBenchmarkMultilingualSTS
984
- config: default
985
- split: test
986
- revision: 29afa2569dcedaaa2fe6a3dcfebab33d28b82e8c
987
- metrics:
988
- - type: cos_sim_pearson
989
- value: 86.42208373352562
990
- - type: cos_sim_spearman
991
- value: 86.99991276887566
992
- - type: euclidean_pearson
993
- value: 85.50325028600815
994
- - type: euclidean_spearman
995
- value: 87.00166758198344
996
- - type: manhattan_pearson
997
- value: 85.51048739822163
998
- - type: manhattan_spearman
999
- value: 86.98373812309134
1000
- - task:
1001
- type: Summarization
1002
- dataset:
1003
- type: lyon-nlp/summarization-summeval-fr-p2p
1004
- name: MTEB SummEvalFr
1005
- config: default
1006
- split: test
1007
- revision: b385812de6a9577b6f4d0f88c6a6e35395a94054
1008
- metrics:
1009
- - type: cos_sim_pearson
1010
- value: 31.911797754639164
1011
- - type: cos_sim_spearman
1012
- value: 32.17186521965941
1013
- - type: dot_pearson
1014
- value: 31.911796813216963
1015
- - type: dot_spearman
1016
- value: 32.17186521965941
1017
- - task:
1018
- type: Reranking
1019
- dataset:
1020
- type: lyon-nlp/mteb-fr-reranking-syntec-s2p
1021
- name: MTEB SyntecReranking
1022
- config: default
1023
- split: test
1024
- revision: daf0863838cd9e3ba50544cdce3ac2b338a1b0ad
1025
- metrics:
1026
- - type: map
1027
- value: 86.04285714285714
1028
- - type: mrr
1029
- value: 86.04285714285714
1030
- - type: nAUC_map_diff1
1031
- value: 67.26948440486838
1032
- - type: nAUC_map_max
1033
- value: 1.8106095852919237
1034
- - type: nAUC_mrr_diff1
1035
- value: 67.26948440486838
1036
- - type: nAUC_mrr_max
1037
- value: 1.8106095852919237
1038
- - task:
1039
- type: Retrieval
1040
- dataset:
1041
- type: lyon-nlp/mteb-fr-retrieval-syntec-s2p
1042
- name: MTEB SyntecRetrieval
1043
- config: default
1044
- split: test
1045
- revision: 19661ccdca4dfc2d15122d776b61685f48c68ca9
1046
- metrics:
1047
- - type: map_at_1
1048
- value: 72.0
1049
- - type: map_at_10
1050
- value: 81.294
1051
- - type: map_at_100
1052
- value: 81.428
1053
- - type: map_at_1000
1054
- value: 81.428
1055
- - type: map_at_20
1056
- value: 81.38499999999999
1057
- - type: map_at_3
1058
- value: 79.833
1059
- - type: map_at_5
1060
- value: 80.88300000000001
1061
- - type: mrr_at_1
1062
- value: 72.0
1063
- - type: mrr_at_10
1064
- value: 81.2940476190476
1065
- - type: mrr_at_100
1066
- value: 81.42843497082626
1067
- - type: mrr_at_1000
1068
- value: 81.42843497082626
1069
- - type: mrr_at_20
1070
- value: 81.3849567099567
1071
- - type: mrr_at_3
1072
- value: 79.83333333333334
1073
- - type: mrr_at_5
1074
- value: 80.88333333333333
1075
- - type: nauc_map_at_1000_diff1
1076
- value: 49.54228699135176
1077
- - type: nauc_map_at_1000_max
1078
- value: 0.06067639126124744
1079
- - type: nauc_map_at_100_diff1
1080
- value: 49.54228699135176
1081
- - type: nauc_map_at_100_max
1082
- value: 0.06067639126124744
1083
- - type: nauc_map_at_10_diff1
1084
- value: 49.53852081026105
1085
- - type: nauc_map_at_10_max
1086
- value: 0.3053366025276881
1087
- - type: nauc_map_at_1_diff1
1088
- value: 46.31028693528697
1089
- - type: nauc_map_at_1_max
1090
- value: -1.3144841269841376
1091
- - type: nauc_map_at_20_diff1
1092
- value: 49.506524393346865
1093
- - type: nauc_map_at_20_max
1094
- value: -0.08620516816116486
1095
- - type: nauc_map_at_3_diff1
1096
- value: 51.19253694434259
1097
- - type: nauc_map_at_3_max
1098
- value: 0.668963341320456
1099
- - type: nauc_map_at_5_diff1
1100
- value: 49.37737335974092
1101
- - type: nauc_map_at_5_max
1102
- value: -0.33590080679527184
1103
- - type: nauc_mrr_at_1000_diff1
1104
- value: 49.54228699135176
1105
- - type: nauc_mrr_at_1000_max
1106
- value: 0.06067639126124744
1107
- - type: nauc_mrr_at_100_diff1
1108
- value: 49.54228699135176
1109
- - type: nauc_mrr_at_100_max
1110
- value: 0.06067639126124744
1111
- - type: nauc_mrr_at_10_diff1
1112
- value: 49.53852081026105
1113
- - type: nauc_mrr_at_10_max
1114
- value: 0.3053366025276881
1115
- - type: nauc_mrr_at_1_diff1
1116
- value: 46.31028693528697
1117
- - type: nauc_mrr_at_1_max
1118
- value: -1.3144841269841376
1119
- - type: nauc_mrr_at_20_diff1
1120
- value: 49.506524393346865
1121
- - type: nauc_mrr_at_20_max
1122
- value: -0.08620516816116486
1123
- - type: nauc_mrr_at_3_diff1
1124
- value: 51.19253694434259
1125
- - type: nauc_mrr_at_3_max
1126
- value: 0.668963341320456
1127
- - type: nauc_mrr_at_5_diff1
1128
- value: 49.37737335974092
1129
- - type: nauc_mrr_at_5_max
1130
- value: -0.33590080679527184
1131
- - type: nauc_ndcg_at_1000_diff1
1132
- value: 49.6848382380357
1133
- - type: nauc_ndcg_at_1000_max
1134
- value: 0.04870501937096382
1135
- - type: nauc_ndcg_at_100_diff1
1136
- value: 49.6848382380357
1137
- - type: nauc_ndcg_at_100_max
1138
- value: 0.04870501937096382
1139
- - type: nauc_ndcg_at_10_diff1
1140
- value: 49.57645777272915
1141
- - type: nauc_ndcg_at_10_max
1142
- value: 0.6430420679440534
1143
- - type: nauc_ndcg_at_1_diff1
1144
- value: 46.31028693528697
1145
- - type: nauc_ndcg_at_1_max
1146
- value: -1.3144841269841376
1147
- - type: nauc_ndcg_at_20_diff1
1148
- value: 49.45017977018584
1149
- - type: nauc_ndcg_at_20_max
1150
- value: -0.9049646537819854
1151
- - type: nauc_ndcg_at_3_diff1
1152
- value: 52.898658060430904
1153
- - type: nauc_ndcg_at_3_max
1154
- value: 1.3070987858400047
1155
- - type: nauc_ndcg_at_5_diff1
1156
- value: 49.054456158711595
1157
- - type: nauc_ndcg_at_5_max
1158
- value: -1.177736876794348
1159
- - type: nauc_precision_at_1000_diff1
1160
- value: nan
1161
- - type: nauc_precision_at_1000_max
1162
- value: nan
1163
- - type: nauc_precision_at_100_diff1
1164
- value: nan
1165
- - type: nauc_precision_at_100_max
1166
- value: nan
1167
- - type: nauc_precision_at_10_diff1
1168
- value: 45.611577964519334
1169
- - type: nauc_precision_at_10_max
1170
- value: 7.936507936508234
1171
- - type: nauc_precision_at_1_diff1
1172
- value: 46.31028693528697
1173
- - type: nauc_precision_at_1_max
1174
- value: -1.3144841269841376
1175
- - type: nauc_precision_at_20_diff1
1176
- value: 35.80765639589114
1177
- - type: nauc_precision_at_20_max
1178
- value: -56.34920634920767
1179
- - type: nauc_precision_at_3_diff1
1180
- value: 61.56395891690006
1181
- - type: nauc_precision_at_3_max
1182
- value: 4.509803921568394
1183
- - type: nauc_precision_at_5_diff1
1184
- value: 43.15592903828254
1185
- - type: nauc_precision_at_5_max
1186
- value: -11.783380018673482
1187
- - type: nauc_recall_at_1000_diff1
1188
- value: nan
1189
- - type: nauc_recall_at_1000_max
1190
- value: nan
1191
- - type: nauc_recall_at_100_diff1
1192
- value: nan
1193
- - type: nauc_recall_at_100_max
1194
- value: nan
1195
- - type: nauc_recall_at_10_diff1
1196
- value: 45.61157796451899
1197
- - type: nauc_recall_at_10_max
1198
- value: 7.9365079365084235
1199
- - type: nauc_recall_at_1_diff1
1200
- value: 46.31028693528697
1201
- - type: nauc_recall_at_1_max
1202
- value: -1.3144841269841376
1203
- - type: nauc_recall_at_20_diff1
1204
- value: 35.80765639589109
1205
- - type: nauc_recall_at_20_max
1206
- value: -56.34920634920657
1207
- - type: nauc_recall_at_3_diff1
1208
- value: 61.5639589169002
1209
- - type: nauc_recall_at_3_max
1210
- value: 4.509803921568655
1211
- - type: nauc_recall_at_5_diff1
1212
- value: 43.15592903828185
1213
- - type: nauc_recall_at_5_max
1214
- value: -11.783380018674132
1215
- - type: ndcg_at_1
1216
- value: 72.0
1217
- - type: ndcg_at_10
1218
- value: 85.39999999999999
1219
- - type: ndcg_at_100
1220
- value: 85.897
1221
- - type: ndcg_at_1000
1222
- value: 85.897
1223
- - type: ndcg_at_20
1224
- value: 85.679
1225
- - type: ndcg_at_3
1226
- value: 82.44
1227
- - type: ndcg_at_5
1228
- value: 84.418
1229
- - type: precision_at_1
1230
- value: 72.0
1231
- - type: precision_at_10
1232
- value: 9.8
1233
- - type: precision_at_100
1234
- value: 1.0
1235
- - type: precision_at_1000
1236
- value: 0.1
1237
- - type: precision_at_20
1238
- value: 4.95
1239
- - type: precision_at_3
1240
- value: 30.0
1241
- - type: precision_at_5
1242
- value: 19.0
1243
- - type: recall_at_1
1244
- value: 72.0
1245
- - type: recall_at_10
1246
- value: 98.0
1247
- - type: recall_at_100
1248
- value: 100.0
1249
- - type: recall_at_1000
1250
- value: 100.0
1251
- - type: recall_at_20
1252
- value: 99.0
1253
- - type: recall_at_3
1254
- value: 90.0
1255
- - type: recall_at_5
1256
- value: 95.0
1257
- - task:
1258
- type: Retrieval
1259
- dataset:
1260
- type: jinaai/xpqa
1261
- name: MTEB XPQARetrieval
1262
- config: default
1263
- split: test
1264
- revision: c99d599f0a6ab9b85b065da6f9d94f9cf731679f
1265
- metrics:
1266
- - type: map_at_1
1267
- value: 39.007999999999996
1268
- - type: map_at_10
1269
- value: 60.319
1270
- - type: map_at_100
1271
- value: 61.644
1272
- - type: map_at_1000
1273
- value: 61.712
1274
- - type: map_at_20
1275
- value: 61.053000000000004
1276
- - type: map_at_3
1277
- value: 53.942
1278
- - type: map_at_5
1279
- value: 58.132
1280
- - type: mrr_at_1
1281
- value: 60.747663551401864
1282
- - type: mrr_at_10
1283
- value: 68.10636404094345
1284
- - type: mrr_at_100
1285
- value: 68.57873500135119
1286
- - type: mrr_at_1000
1287
- value: 68.60183171580495
1288
- - type: mrr_at_20
1289
- value: 68.36478690417064
1290
- - type: mrr_at_3
1291
- value: 66.08811748998662
1292
- - type: mrr_at_5
1293
- value: 67.12950600801062
1294
- - type: nauc_map_at_1000_diff1
1295
- value: 46.633293504114434
1296
- - type: nauc_map_at_1000_max
1297
- value: 49.64101214126472
1298
- - type: nauc_map_at_100_diff1
1299
- value: 46.591074735810764
1300
- - type: nauc_map_at_100_max
1301
- value: 49.60435205919251
1302
- - type: nauc_map_at_10_diff1
1303
- value: 46.43486044009182
1304
- - type: nauc_map_at_10_max
1305
- value: 49.222376322201065
1306
- - type: nauc_map_at_1_diff1
1307
- value: 54.05448738315762
1308
- - type: nauc_map_at_1_max
1309
- value: 24.89423418246206
1310
- - type: nauc_map_at_20_diff1
1311
- value: 46.5131815367993
1312
- - type: nauc_map_at_20_max
1313
- value: 49.59934686413147
1314
- - type: nauc_map_at_3_diff1
1315
- value: 48.9688624037045
1316
- - type: nauc_map_at_3_max
1317
- value: 42.49957358403678
1318
- - type: nauc_map_at_5_diff1
1319
- value: 46.99183039261338
1320
- - type: nauc_map_at_5_max
1321
- value: 47.53639129265315
1322
- - type: nauc_mrr_at_1000_diff1
1323
- value: 55.19161872624107
1324
- - type: nauc_mrr_at_1000_max
1325
- value: 57.00358990989949
1326
- - type: nauc_mrr_at_100_diff1
1327
- value: 55.18922613003231
1328
- - type: nauc_mrr_at_100_max
1329
- value: 56.99910514727928
1330
- - type: nauc_mrr_at_10_diff1
1331
- value: 55.030904237371224
1332
- - type: nauc_mrr_at_10_max
1333
- value: 56.99207425207498
1334
- - type: nauc_mrr_at_1_diff1
1335
- value: 57.34582863547949
1336
- - type: nauc_mrr_at_1_max
1337
- value: 56.98349812853321
1338
- - type: nauc_mrr_at_20_diff1
1339
- value: 55.0844531407365
1340
- - type: nauc_mrr_at_20_max
1341
- value: 57.00264208604279
1342
- - type: nauc_mrr_at_3_diff1
1343
- value: 55.64735827895618
1344
- - type: nauc_mrr_at_3_max
1345
- value: 57.29703659670222
1346
- - type: nauc_mrr_at_5_diff1
1347
- value: 55.38932070005733
1348
- - type: nauc_mrr_at_5_max
1349
- value: 57.25407452051235
1350
- - type: nauc_ndcg_at_1000_diff1
1351
- value: 48.35929794131023
1352
- - type: nauc_ndcg_at_1000_max
1353
- value: 52.522778834890325
1354
- - type: nauc_ndcg_at_100_diff1
1355
- value: 47.56861995950162
1356
- - type: nauc_ndcg_at_100_max
1357
- value: 51.88529740719871
1358
- - type: nauc_ndcg_at_10_diff1
1359
- value: 46.87985909159253
1360
- - type: nauc_ndcg_at_10_max
1361
- value: 51.38385177949589
1362
- - type: nauc_ndcg_at_1_diff1
1363
- value: 57.34582863547949
1364
- - type: nauc_ndcg_at_1_max
1365
- value: 56.98349812853321
1366
- - type: nauc_ndcg_at_20_diff1
1367
- value: 47.02262106267654
1368
- - type: nauc_ndcg_at_20_max
1369
- value: 51.98897867452814
1370
- - type: nauc_ndcg_at_3_diff1
1371
- value: 49.03406876232345
1372
- - type: nauc_ndcg_at_3_max
1373
- value: 51.51024175391598
1374
- - type: nauc_ndcg_at_5_diff1
1375
- value: 47.91197516404686
1376
- - type: nauc_ndcg_at_5_max
1377
- value: 49.81634328349405
1378
- - type: nauc_precision_at_1000_diff1
1379
- value: -17.71701381388244
1380
- - type: nauc_precision_at_1000_max
1381
- value: 21.388531798131364
1382
- - type: nauc_precision_at_100_diff1
1383
- value: -15.539336210367946
1384
- - type: nauc_precision_at_100_max
1385
- value: 24.698569356368356
1386
- - type: nauc_precision_at_10_diff1
1387
- value: -4.847564718245167
1388
- - type: nauc_precision_at_10_max
1389
- value: 36.89232193362945
1390
- - type: nauc_precision_at_1_diff1
1391
- value: 57.34582863547949
1392
- - type: nauc_precision_at_1_max
1393
- value: 56.98349812853321
1394
- - type: nauc_precision_at_20_diff1
1395
- value: -8.665616069723095
1396
- - type: nauc_precision_at_20_max
1397
- value: 33.10610080847679
1398
- - type: nauc_precision_at_3_diff1
1399
- value: 11.05940284130611
1400
- - type: nauc_precision_at_3_max
1401
- value: 46.70869419036014
1402
- - type: nauc_precision_at_5_diff1
1403
- value: 1.5834393191629197
1404
- - type: nauc_precision_at_5_max
1405
- value: 41.862527913783865
1406
- - type: nauc_recall_at_1000_diff1
1407
- value: -10.569638499832076
1408
- - type: nauc_recall_at_1000_max
1409
- value: 8.749844537815326
1410
- - type: nauc_recall_at_100_diff1
1411
- value: 27.487071443233262
1412
- - type: nauc_recall_at_100_max
1413
- value: 35.4122293490153
1414
- - type: nauc_recall_at_10_diff1
1415
- value: 35.91432645476102
1416
- - type: nauc_recall_at_10_max
1417
- value: 44.19204765665408
1418
- - type: nauc_recall_at_1_diff1
1419
- value: 54.05448738315762
1420
- - type: nauc_recall_at_1_max
1421
- value: 24.89423418246206
1422
- - type: nauc_recall_at_20_diff1
1423
- value: 34.4589376419412
1424
- - type: nauc_recall_at_20_max
1425
- value: 44.798804712797136
1426
- - type: nauc_recall_at_3_diff1
1427
- value: 45.60515874084114
1428
- - type: nauc_recall_at_3_max
1429
- value: 39.109212749550764
1430
- - type: nauc_recall_at_5_diff1
1431
- value: 41.35313462606428
1432
- - type: nauc_recall_at_5_max
1433
- value: 43.39841866434438
1434
- - type: ndcg_at_1
1435
- value: 60.748000000000005
1436
- - type: ndcg_at_10
1437
- value: 66.232
1438
- - type: ndcg_at_100
1439
- value: 70.745
1440
- - type: ndcg_at_1000
1441
- value: 71.87
1442
- - type: ndcg_at_20
1443
- value: 67.977
1444
- - type: ndcg_at_3
1445
- value: 61.06399999999999
1446
- - type: ndcg_at_5
1447
- value: 62.744
1448
- - type: precision_at_1
1449
- value: 60.748000000000005
1450
- - type: precision_at_10
1451
- value: 15.434000000000001
1452
- - type: precision_at_100
1453
- value: 1.924
1454
- - type: precision_at_1000
1455
- value: 0.207
1456
- - type: precision_at_20
1457
- value: 8.378
1458
- - type: precision_at_3
1459
- value: 37.161
1460
- - type: precision_at_5
1461
- value: 26.916
1462
- - type: recall_at_1
1463
- value: 39.007999999999996
1464
- - type: recall_at_10
1465
- value: 74.929
1466
- - type: recall_at_100
1467
- value: 92.508
1468
- - type: recall_at_1000
1469
- value: 99.8
1470
- - type: recall_at_20
1471
- value: 80.343
1472
- - type: recall_at_3
1473
- value: 58.367000000000004
1474
- - type: recall_at_5
1475
- value: 66.446
1476
- license: apache-2.0
1477
- language:
1478
- - fr
1479
- - en
1480
- ---
1481
-
1482
- # [bilingual-embedding-large](https://huggingface.co/Lajavaness/bilingual-embedding-large)
1483
-
1484
- 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 [XLM-RoBERTa](https://huggingface.co/FacebookAI/xlm-roberta-large), a pre-trained language model based on the [XLM-RoBERTa](https://huggingface.co/FacebookAI/xlm-roberta-large) 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.
1485
-
1486
-
1487
- ## Full Model Architecture
1488
- ```
1489
- SentenceTransformer(
1490
- (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BilingualModel
1491
- (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})
1492
- (2): Normalize()
1493
- )
1494
- ```
1495
-
1496
- ## Training and Fine-tuning process
1497
- #### Stage 1: NLI Training
1498
- - Dataset: [(SNLI+XNLI) for english+french]
1499
- - 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.
1500
- ### Stage 3: Continued Fine-tuning for Semantic Textual Similarity on STS Benchmark
1501
- - Dataset: [STSB-fr and en]
1502
- - Method: Fine-tuning specifically for the semantic textual similarity benchmark using Siamese BERT-Networks configured with the 'sentence-transformers' library.
1503
- ### Stage 4: Advanced Augmentation Fine-tuning
1504
- - Dataset: STSB-vn with generate [silver sample from gold sample](https://www.sbert.net/examples/training/data_augmentation/README.html)
1505
- - 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.
1506
-
1507
-
1508
- ## Usage:
1509
-
1510
- Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
1511
-
1512
- ```
1513
- pip install -U sentence-transformers
1514
- ```
1515
-
1516
- Then you can use the model like this:
1517
-
1518
- ```python
1519
- from sentence_transformers import SentenceTransformer
1520
- from pyvi.ViTokenizer import tokenize
1521
-
1522
- sentences = ["Paris est une capitale de la France", "Paris is a capital of France"]
1523
-
1524
- model = SentenceTransformer('Lajavaness/bilingual-embedding-large', trust_remote_code=True)
1525
- print(embeddings)
1526
-
1527
- ```
1528
-
1529
-
1530
-
1531
-
1532
-
1533
- ## Evaluation
1534
-
1535
- TODO
1536
-
1537
- ## Citation
1538
-
1539
- @article{conneau2019unsupervised,
1540
- title={Unsupervised cross-lingual representation learning at scale},
1541
- 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},
1542
- journal={arXiv preprint arXiv:1911.02116},
1543
- year={2019}
1544
- }
1545
-
1546
- @article{reimers2019sentence,
1547
- title={Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks},
1548
- author={Nils Reimers, Iryna Gurevych},
1549
- journal={https://arxiv.org/abs/1908.10084},
1550
- year={2019}
1551
- }
1552
-
1553
- @article{thakur2020augmented,
1554
- title={Augmented SBERT: Data Augmentation Method for Improving Bi-Encoders for Pairwise Sentence Scoring Tasks},
1555
- author={Thakur, Nandan and Reimers, Nils and Daxenberger, Johannes and Gurevych, Iryna},
1556
- journal={arXiv e-prints},
1557
- pages={arXiv--2010},
1558
- year={2020}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:503b71b7898475dbd4e4b16e1f7483ce1989faa17e71920c7ec5af4e90d34be0
3
  size 2239607176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39915c4e22d7ef7b7a68aaa59da6e132d16900d8e0696845578fe7e334360a25
3
  size 2239607176