biswa921 commited on
Commit
49c73ba
1 Parent(s): 6df18c6

Mteb benchmark

Browse files
Files changed (1) hide show
  1. README.md +506 -0
README.md ADDED
@@ -0,0 +1,506 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - mteb
4
+ model-index:
5
+ - name: bge-m3
6
+ results:
7
+ - task:
8
+ type: Classification
9
+ dataset:
10
+ type: mteb/amazon_counterfactual
11
+ name: MTEB AmazonCounterfactualClassification (en)
12
+ config: en
13
+ split: test
14
+ revision: e8379541af4e31359cca9fbcf4b00f2671dba205
15
+ metrics:
16
+ - type: accuracy
17
+ value: 75.6268656716418
18
+ - type: ap
19
+ value: 39.50276109614102
20
+ - type: f1
21
+ value: 70.00224623431103
22
+ - task:
23
+ type: Clustering
24
+ dataset:
25
+ type: mteb/arxiv-clustering-p2p
26
+ name: MTEB ArxivClusteringP2P
27
+ config: default
28
+ split: test
29
+ revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d
30
+ metrics:
31
+ - type: v_measure
32
+ value: 39.409674498704625
33
+ - task:
34
+ type: Reranking
35
+ dataset:
36
+ type: mteb/askubuntudupquestions-reranking
37
+ name: MTEB AskUbuntuDupQuestions
38
+ config: default
39
+ split: test
40
+ revision: 2000358ca161889fa9c082cb41daa8dcfb161a54
41
+ metrics:
42
+ - type: map
43
+ value: 61.52757354203137
44
+ - type: mrr
45
+ value: 74.28241656773513
46
+ - task:
47
+ type: STS
48
+ dataset:
49
+ type: mteb/biosses-sts
50
+ name: MTEB BIOSSES
51
+ config: default
52
+ split: test
53
+ revision: d3fb88f8f02e40887cd149695127462bbcf29b4a
54
+ metrics:
55
+ - type: cos_sim_pearson
56
+ value: 84.39442490594014
57
+ - type: cos_sim_spearman
58
+ value: 83.37599616417513
59
+ - type: euclidean_pearson
60
+ value: 83.23317790460271
61
+ - type: euclidean_spearman
62
+ value: 83.37599616417513
63
+ - type: manhattan_pearson
64
+ value: 83.23182214744224
65
+ - type: manhattan_spearman
66
+ value: 83.5428674363298
67
+ - task:
68
+ type: Classification
69
+ dataset:
70
+ type: mteb/banking77
71
+ name: MTEB Banking77Classification
72
+ config: default
73
+ split: test
74
+ revision: 0fd18e25b25c072e09e0d92ab615fda904d66300
75
+ metrics:
76
+ - type: accuracy
77
+ value: 81.93181818181819
78
+ - type: f1
79
+ value: 81.0852312152688
80
+ - task:
81
+ type: Classification
82
+ dataset:
83
+ type: mteb/emotion
84
+ name: MTEB EmotionClassification
85
+ config: default
86
+ split: test
87
+ revision: 4f58c6b202a23cf9a4da393831edf4f9183cad37
88
+ metrics:
89
+ - type: accuracy
90
+ value: 50.16499999999999
91
+ - type: f1
92
+ value: 43.57906972116264
93
+ - task:
94
+ type: Classification
95
+ dataset:
96
+ type: mteb/mtop_domain
97
+ name: MTEB MTOPDomainClassification (en)
98
+ config: en
99
+ split: test
100
+ revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf
101
+ metrics:
102
+ - type: accuracy
103
+ value: 93.35841313269493
104
+ - type: f1
105
+ value: 93.060022693275
106
+ - task:
107
+ type: Classification
108
+ dataset:
109
+ type: mteb/mtop_intent
110
+ name: MTEB MTOPIntentClassification (en)
111
+ config: en
112
+ split: test
113
+ revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba
114
+ metrics:
115
+ - type: accuracy
116
+ value: 66.58002735978113
117
+ - type: f1
118
+ value: 46.995919480823055
119
+ - task:
120
+ type: Classification
121
+ dataset:
122
+ type: mteb/amazon_massive_intent
123
+ name: MTEB MassiveIntentClassification (en)
124
+ config: en
125
+ split: test
126
+ revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7
127
+ metrics:
128
+ - type: accuracy
129
+ value: 71.07935440484196
130
+ - type: f1
131
+ value: 69.13197875645403
132
+ - task:
133
+ type: Classification
134
+ dataset:
135
+ type: mteb/amazon_massive_scenario
136
+ name: MTEB MassiveScenarioClassification (en)
137
+ config: en
138
+ split: test
139
+ revision: 7d571f92784cd94a019292a1f45445077d0ef634
140
+ metrics:
141
+ - type: accuracy
142
+ value: 76.63752521856087
143
+ - type: f1
144
+ value: 75.61348469613843
145
+ - task:
146
+ type: STS
147
+ dataset:
148
+ type: mteb/sickr-sts
149
+ name: MTEB SICK-R
150
+ config: default
151
+ split: test
152
+ revision: a6ea5a8cab320b040a23452cc28066d9beae2cee
153
+ metrics:
154
+ - type: cos_sim_pearson
155
+ value: 83.33269306539026
156
+ - type: cos_sim_spearman
157
+ value: 79.71441518631086
158
+ - type: euclidean_pearson
159
+ value: 80.98109404189279
160
+ - type: euclidean_spearman
161
+ value: 79.71444969096095
162
+ - type: manhattan_pearson
163
+ value: 80.97223989357175
164
+ - type: manhattan_spearman
165
+ value: 79.64929261210406
166
+ - task:
167
+ type: STS
168
+ dataset:
169
+ type: mteb/sts12-sts
170
+ name: MTEB STS12
171
+ config: default
172
+ split: test
173
+ revision: a0d554a64d88156834ff5ae9920b964011b16384
174
+ metrics:
175
+ - type: cos_sim_pearson
176
+ value: 83.7127498314437
177
+ - type: cos_sim_spearman
178
+ value: 78.73426610516154
179
+ - type: euclidean_pearson
180
+ value: 79.72827173736742
181
+ - type: euclidean_spearman
182
+ value: 78.731973450314
183
+ - type: manhattan_pearson
184
+ value: 79.71391822179304
185
+ - type: manhattan_spearman
186
+ value: 78.69626503719782
187
+ - task:
188
+ type: STS
189
+ dataset:
190
+ type: mteb/sts13-sts
191
+ name: MTEB STS13
192
+ config: default
193
+ split: test
194
+ revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca
195
+ metrics:
196
+ - type: cos_sim_pearson
197
+ value: 78.33449726355023
198
+ - type: cos_sim_spearman
199
+ value: 79.59703323420547
200
+ - type: euclidean_pearson
201
+ value: 79.87238808505464
202
+ - type: euclidean_spearman
203
+ value: 79.59703323420547
204
+ - type: manhattan_pearson
205
+ value: 79.5006260085966
206
+ - type: manhattan_spearman
207
+ value: 79.21864659717262
208
+ - task:
209
+ type: STS
210
+ dataset:
211
+ type: mteb/sts14-sts
212
+ name: MTEB STS14
213
+ config: default
214
+ split: test
215
+ revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375
216
+ metrics:
217
+ - type: cos_sim_pearson
218
+ value: 79.00088445445654
219
+ - type: cos_sim_spearman
220
+ value: 78.99977508575147
221
+ - type: euclidean_pearson
222
+ value: 78.63222924140206
223
+ - type: euclidean_spearman
224
+ value: 78.99976994069327
225
+ - type: manhattan_pearson
226
+ value: 78.35504771673297
227
+ - type: manhattan_spearman
228
+ value: 78.76306077740067
229
+ - task:
230
+ type: STS
231
+ dataset:
232
+ type: mteb/sts15-sts
233
+ name: MTEB STS15
234
+ config: default
235
+ split: test
236
+ revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3
237
+ metrics:
238
+ - type: cos_sim_pearson
239
+ value: 87.13160613452308
240
+ - type: cos_sim_spearman
241
+ value: 87.81435104273643
242
+ - type: euclidean_pearson
243
+ value: 87.22395745487297
244
+ - type: euclidean_spearman
245
+ value: 87.81435041827874
246
+ - type: manhattan_pearson
247
+ value: 87.17630476262896
248
+ - type: manhattan_spearman
249
+ value: 87.76535338976686
250
+ - task:
251
+ type: STS
252
+ dataset:
253
+ type: mteb/sts16-sts
254
+ name: MTEB STS16
255
+ config: default
256
+ split: test
257
+ revision: 4d8694f8f0e0100860b497b999b3dbed754a0513
258
+ metrics:
259
+ - type: cos_sim_pearson
260
+ value: 83.76424652225954
261
+ - type: cos_sim_spearman
262
+ value: 85.39745570134193
263
+ - type: euclidean_pearson
264
+ value: 84.6971466556576
265
+ - type: euclidean_spearman
266
+ value: 85.39745570134193
267
+ - type: manhattan_pearson
268
+ value: 84.61210275324463
269
+ - type: manhattan_spearman
270
+ value: 85.30727114432379
271
+ - task:
272
+ type: STS
273
+ dataset:
274
+ type: mteb/sts17-crosslingual-sts
275
+ name: MTEB STS17 (en-en)
276
+ config: en-en
277
+ split: test
278
+ revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d
279
+ metrics:
280
+ - type: cos_sim_pearson
281
+ value: 86.87956530541486
282
+ - type: cos_sim_spearman
283
+ value: 87.13412608536781
284
+ - type: euclidean_pearson
285
+ value: 87.80084186244981
286
+ - type: euclidean_spearman
287
+ value: 87.13412608536781
288
+ - type: manhattan_pearson
289
+ value: 87.73101535306475
290
+ - type: manhattan_spearman
291
+ value: 87.05897655963285
292
+ - task:
293
+ type: STS
294
+ dataset:
295
+ type: mteb/stsbenchmark-sts
296
+ name: MTEB STSBenchmark
297
+ config: default
298
+ split: test
299
+ revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831
300
+ metrics:
301
+ - type: cos_sim_pearson
302
+ value: 83.70737517925419
303
+ - type: cos_sim_spearman
304
+ value: 84.84687698325351
305
+ - type: euclidean_pearson
306
+ value: 84.36525309890885
307
+ - type: euclidean_spearman
308
+ value: 84.84688249844098
309
+ - type: manhattan_pearson
310
+ value: 84.31171573973266
311
+ - type: manhattan_spearman
312
+ value: 84.79550448196474
313
+ - task:
314
+ type: PairClassification
315
+ dataset:
316
+ type: mteb/sprintduplicatequestions-pairclassification
317
+ name: MTEB SprintDuplicateQuestions
318
+ config: default
319
+ split: test
320
+ revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46
321
+ metrics:
322
+ - type: cos_sim_accuracy
323
+ value: 99.87722772277228
324
+ - type: cos_sim_ap
325
+ value: 97.32479581402639
326
+ - type: cos_sim_f1
327
+ value: 93.74369323915236
328
+ - type: cos_sim_precision
329
+ value: 94.60285132382892
330
+ - type: cos_sim_recall
331
+ value: 92.9
332
+ - type: dot_accuracy
333
+ value: 99.87722772277228
334
+ - type: dot_ap
335
+ value: 97.32479581402637
336
+ - type: dot_f1
337
+ value: 93.74369323915236
338
+ - type: dot_precision
339
+ value: 94.60285132382892
340
+ - type: dot_recall
341
+ value: 92.9
342
+ - type: euclidean_accuracy
343
+ value: 99.87722772277228
344
+ - type: euclidean_ap
345
+ value: 97.32479581402639
346
+ - type: euclidean_f1
347
+ value: 93.74369323915236
348
+ - type: euclidean_precision
349
+ value: 94.60285132382892
350
+ - type: euclidean_recall
351
+ value: 92.9
352
+ - type: manhattan_accuracy
353
+ value: 99.87524752475248
354
+ - type: manhattan_ap
355
+ value: 97.29133330261223
356
+ - type: manhattan_f1
357
+ value: 93.59359359359361
358
+ - type: manhattan_precision
359
+ value: 93.687374749499
360
+ - type: manhattan_recall
361
+ value: 93.5
362
+ - type: max_accuracy
363
+ value: 99.87722772277228
364
+ - type: max_ap
365
+ value: 97.32479581402639
366
+ - type: max_f1
367
+ value: 93.74369323915236
368
+ - task:
369
+ type: Classification
370
+ dataset:
371
+ type: mteb/toxic_conversations_50k
372
+ name: MTEB ToxicConversationsClassification
373
+ config: default
374
+ split: test
375
+ revision: d7c0de2777da35d6aae2200a62c6e0e5af397c4c
376
+ metrics:
377
+ - type: accuracy
378
+ value: 72.60060000000001
379
+ - type: ap
380
+ value: 15.719924742317021
381
+ - type: f1
382
+ value: 56.30561683159878
383
+ - task:
384
+ type: Classification
385
+ dataset:
386
+ type: mteb/tweet_sentiment_extraction
387
+ name: MTEB TweetSentimentExtractionClassification
388
+ config: default
389
+ split: test
390
+ revision: d604517c81ca91fe16a244d1248fc021f9ecee7a
391
+ metrics:
392
+ - type: accuracy
393
+ value: 63.71250707413696
394
+ - type: f1
395
+ value: 63.54808116265952
396
+ - task:
397
+ type: PairClassification
398
+ dataset:
399
+ type: mteb/twittersemeval2015-pairclassification
400
+ name: MTEB TwitterSemEval2015
401
+ config: default
402
+ split: test
403
+ revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1
404
+ metrics:
405
+ - type: cos_sim_accuracy
406
+ value: 85.110568039578
407
+ - type: cos_sim_ap
408
+ value: 70.28927714315245
409
+ - type: cos_sim_f1
410
+ value: 65.03893361488716
411
+ - type: cos_sim_precision
412
+ value: 65.06469500924214
413
+ - type: cos_sim_recall
414
+ value: 65.0131926121372
415
+ - type: dot_accuracy
416
+ value: 85.110568039578
417
+ - type: dot_ap
418
+ value: 70.28928082939848
419
+ - type: dot_f1
420
+ value: 65.03893361488716
421
+ - type: dot_precision
422
+ value: 65.06469500924214
423
+ - type: dot_recall
424
+ value: 65.0131926121372
425
+ - type: euclidean_accuracy
426
+ value: 85.110568039578
427
+ - type: euclidean_ap
428
+ value: 70.28928621260852
429
+ - type: euclidean_f1
430
+ value: 65.03893361488716
431
+ - type: euclidean_precision
432
+ value: 65.06469500924214
433
+ - type: euclidean_recall
434
+ value: 65.0131926121372
435
+ - type: manhattan_accuracy
436
+ value: 85.02115992132086
437
+ - type: manhattan_ap
438
+ value: 70.05813255171925
439
+ - type: manhattan_f1
440
+ value: 64.59658311510164
441
+ - type: manhattan_precision
442
+ value: 61.24379285883188
443
+ - type: manhattan_recall
444
+ value: 68.33773087071239
445
+ - type: max_accuracy
446
+ value: 85.110568039578
447
+ - type: max_ap
448
+ value: 70.28928621260852
449
+ - type: max_f1
450
+ value: 65.03893361488716
451
+ - task:
452
+ type: PairClassification
453
+ dataset:
454
+ type: mteb/twitterurlcorpus-pairclassification
455
+ name: MTEB TwitterURLCorpus
456
+ config: default
457
+ split: test
458
+ revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf
459
+ metrics:
460
+ - type: cos_sim_accuracy
461
+ value: 88.99949547871309
462
+ - type: cos_sim_ap
463
+ value: 85.82819569154559
464
+ - type: cos_sim_f1
465
+ value: 78.37315338318439
466
+ - type: cos_sim_precision
467
+ value: 74.46454564358494
468
+ - type: cos_sim_recall
469
+ value: 82.71481367416075
470
+ - type: dot_accuracy
471
+ value: 88.99949547871309
472
+ - type: dot_ap
473
+ value: 85.82820043407936
474
+ - type: dot_f1
475
+ value: 78.37315338318439
476
+ - type: dot_precision
477
+ value: 74.46454564358494
478
+ - type: dot_recall
479
+ value: 82.71481367416075
480
+ - type: euclidean_accuracy
481
+ value: 88.99949547871309
482
+ - type: euclidean_ap
483
+ value: 85.82819622588083
484
+ - type: euclidean_f1
485
+ value: 78.37315338318439
486
+ - type: euclidean_precision
487
+ value: 74.46454564358494
488
+ - type: euclidean_recall
489
+ value: 82.71481367416075
490
+ - type: manhattan_accuracy
491
+ value: 88.98009081383165
492
+ - type: manhattan_ap
493
+ value: 85.77393389750326
494
+ - type: manhattan_f1
495
+ value: 78.38852097130243
496
+ - type: manhattan_precision
497
+ value: 75.06341600901916
498
+ - type: manhattan_recall
499
+ value: 82.0218663381583
500
+ - type: max_accuracy
501
+ value: 88.99949547871309
502
+ - type: max_ap
503
+ value: 85.82820043407936
504
+ - type: max_f1
505
+ value: 78.38852097130243
506
+ ---