Datasets:
mteb
/

orionweller commited on
Commit
46c0fca
0 Parent(s):
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .github/workflows/main.yml +19 -0
  2. CONTRIBUTING +3 -0
  3. README.md +5 -0
  4. paths.json +0 -0
  5. remove_spaces_from_large_json_files.py +15 -0
  6. results.py +369 -0
  7. results/Baichuan-text-embedding/no_revision_available/AFQMC.json +20 -0
  8. results/Baichuan-text-embedding/no_revision_available/ATEC.json +20 -0
  9. results/Baichuan-text-embedding/no_revision_available/AmazonReviewsClassification.json +25 -0
  10. results/Baichuan-text-embedding/no_revision_available/BQ.json +20 -0
  11. results/Baichuan-text-embedding/no_revision_available/CLSClusteringP2P.json +10 -0
  12. results/Baichuan-text-embedding/no_revision_available/CLSClusteringS2S.json +10 -0
  13. results/Baichuan-text-embedding/no_revision_available/CMedQAv1.json +10 -0
  14. results/Baichuan-text-embedding/no_revision_available/CMedQAv2.json +10 -0
  15. results/Baichuan-text-embedding/no_revision_available/CmedqaRetrieval.json +38 -0
  16. results/Baichuan-text-embedding/no_revision_available/Cmnli.json +49 -0
  17. results/Baichuan-text-embedding/no_revision_available/CovidRetrieval.json +38 -0
  18. results/Baichuan-text-embedding/no_revision_available/DuRetrieval.json +38 -0
  19. results/Baichuan-text-embedding/no_revision_available/EcomRetrieval.json +38 -0
  20. results/Baichuan-text-embedding/no_revision_available/IFlyTek.json +13 -0
  21. results/Baichuan-text-embedding/no_revision_available/JDReview.json +15 -0
  22. results/Baichuan-text-embedding/no_revision_available/LCQMC.json +20 -0
  23. results/Baichuan-text-embedding/no_revision_available/MMarcoReranking.json +10 -0
  24. results/Baichuan-text-embedding/no_revision_available/MMarcoRetrieval.json +38 -0
  25. results/Baichuan-text-embedding/no_revision_available/MassiveIntentClassification.json +25 -0
  26. results/Baichuan-text-embedding/no_revision_available/MassiveScenarioClassification.json +25 -0
  27. results/Baichuan-text-embedding/no_revision_available/MedicalRetrieval.json +38 -0
  28. results/Baichuan-text-embedding/no_revision_available/MultilingualSentiment.json +13 -0
  29. results/Baichuan-text-embedding/no_revision_available/Ocnli.json +49 -0
  30. results/Baichuan-text-embedding/no_revision_available/OnlineShopping.json +15 -0
  31. results/Baichuan-text-embedding/no_revision_available/PAWSX.json +20 -0
  32. results/Baichuan-text-embedding/no_revision_available/QBQTC.json +20 -0
  33. results/Baichuan-text-embedding/no_revision_available/STS22.json +22 -0
  34. results/Baichuan-text-embedding/no_revision_available/STSB.json +20 -0
  35. results/Baichuan-text-embedding/no_revision_available/T2Reranking.json +10 -0
  36. results/Baichuan-text-embedding/no_revision_available/T2Retrieval.json +38 -0
  37. results/Baichuan-text-embedding/no_revision_available/TNews.json +13 -0
  38. results/Baichuan-text-embedding/no_revision_available/ThuNewsClusteringP2P.json +10 -0
  39. results/Baichuan-text-embedding/no_revision_available/ThuNewsClusteringS2S.json +10 -0
  40. results/Baichuan-text-embedding/no_revision_available/VideoRetrieval.json +38 -0
  41. results/Baichuan-text-embedding/no_revision_available/Waimai.json +15 -0
  42. results/Cohere-embed-english-v3.0/no_revision_available/AILACasedocs.json +1 -0
  43. results/Cohere-embed-english-v3.0/no_revision_available/AILAStatutes.json +1 -0
  44. results/Cohere-embed-english-v3.0/no_revision_available/Core17InstructionRetrieval.json +9 -0
  45. results/Cohere-embed-english-v3.0/no_revision_available/GerDaLIRSmall.json +1 -0
  46. results/Cohere-embed-english-v3.0/no_revision_available/LeCaRDv2.json +1 -0
  47. results/Cohere-embed-english-v3.0/no_revision_available/LegalBenchConsumerContractsQA.json +1 -0
  48. results/Cohere-embed-english-v3.0/no_revision_available/LegalBenchCorporateLobbying.json +1 -0
  49. results/Cohere-embed-english-v3.0/no_revision_available/LegalQuAD.json +1 -0
  50. results/Cohere-embed-english-v3.0/no_revision_available/LegalSummarization.json +1 -0
.github/workflows/main.yml ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: Sync to Hugging Face hub
2
+ on:
3
+ push:
4
+ branches: [main]
5
+ # to run this workflow manually from the Actions tab
6
+ workflow_dispatch:
7
+
8
+ jobs:
9
+ sync-to-hub:
10
+ runs-on: ubuntu-latest
11
+ steps:
12
+ - uses: actions/checkout@v3
13
+ with:
14
+ fetch-depth: 0
15
+ lfs: true
16
+ - name: Push to hub
17
+ env:
18
+ HF_TOKEN: ${{ secrets.HF_TOKEN }}
19
+ run: git push https://mteb:[email protected]/datasets/mteb/results main
CONTRIBUTING ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ TODO
2
+
3
+ Users must be sure no files are over 10MB. If there are we should remove all spaces from them to keep them < 10MB. This should be a pre-commit hook checking.
README.md ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ ---
2
+ benchmark: mteb
3
+ type: evaluation
4
+ submission_name: MTEB
5
+ ---
paths.json ADDED
The diff for this file is too large to render. See raw diff
 
remove_spaces_from_large_json_files.py ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+ import glob
3
+ import sys
4
+ import json
5
+
6
+ for file in glob.glob("results/*/*/*.json"):
7
+ # if the file is greater than 9 MB, compress it with gzip
8
+ if os.path.getsize(file) >= 9.5 * 1024 * 1024:
9
+ print(f"Resizing {file} to have no indentations")
10
+ # read it in as json and write it out with no indent
11
+ with open(file, "r") as f:
12
+ data = json.load(f)
13
+
14
+ with open(file, "w") as f:
15
+ json.dump(data, f, indent=None)
results.py ADDED
@@ -0,0 +1,369 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """MTEB Results"""
2
+
3
+ import json
4
+
5
+ import datasets
6
+
7
+
8
+ logger = datasets.logging.get_logger(__name__)
9
+
10
+
11
+ _CITATION = """@article{muennighoff2022mteb,
12
+ doi = {10.48550/ARXIV.2210.07316},
13
+ url = {https://arxiv.org/abs/2210.07316},
14
+ author = {Muennighoff, Niklas and Tazi, Nouamane and Magne, Lo{\"\i}c and Reimers, Nils},
15
+ title = {MTEB: Massive Text Embedding Benchmark},
16
+ publisher = {arXiv},
17
+ journal={arXiv preprint arXiv:2210.07316},
18
+ year = {2022}
19
+ }
20
+ """
21
+
22
+ _DESCRIPTION = """Results on MTEB"""
23
+
24
+ URL = "https://huggingface.co/datasets/mteb/results/resolve/main/paths.json"
25
+ VERSION = datasets.Version("1.0.1")
26
+ EVAL_LANGS = ['af', 'afr-eng', 'am', "amh", 'amh-eng', 'ang-eng', 'ar', 'ar-ar', 'ara-eng', 'arq-eng', 'arz-eng', 'ast-eng', 'awa-eng', 'az', 'aze-eng', 'bel-eng', 'ben-eng', 'ber-eng', 'bn', 'bos-eng', 'bre-eng', 'bul-eng', 'cat-eng', 'cbk-eng', 'ceb-eng', 'ces-eng', 'cha-eng', 'cmn-eng', 'cor-eng', 'csb-eng', 'cy', 'cym-eng', 'da', 'dan-eng', 'de', 'de-fr', 'de-pl', 'deu-eng', 'dsb-eng', 'dtp-eng', 'el', 'ell-eng', 'en', 'en-ar', 'en-de', 'en-en', 'en-tr', 'eng', 'epo-eng', 'es', 'es-en', 'es-es', 'es-it', 'est-eng', 'eus-eng', 'fa', 'fao-eng', 'fi', 'fin-eng', 'fr', 'fr-en', 'fr-pl', 'fra', 'fra-eng', 'fry-eng', 'gla-eng', 'gle-eng', 'glg-eng', 'gsw-eng', 'hau', 'he', 'heb-eng', 'hi', 'hin-eng', 'hrv-eng', 'hsb-eng', 'hu', 'hun-eng', 'hy', 'hye-eng', 'ibo', 'id', 'ido-eng', 'ile-eng', 'ina-eng', 'ind-eng', 'is', 'isl-eng', 'it', 'it-en', 'ita-eng', 'ja', 'jav-eng', 'jpn-eng', 'jv', 'ka', 'kab-eng', 'kat-eng', 'kaz-eng', 'khm-eng', 'km', 'kn', 'ko', 'ko-ko', 'kor-eng', 'kur-eng', 'kzj-eng', 'lat-eng', 'lfn-eng', 'lit-eng', 'lin', 'lug', 'lv', 'lvs-eng', 'mal-eng', 'mar-eng', 'max-eng', 'mhr-eng', 'mkd-eng', 'ml', 'mn', 'mon-eng', 'ms', 'my', 'nb', 'nds-eng', 'nl', 'nl-ende-en', 'nld-eng', 'nno-eng', 'nob-eng', 'nov-eng', 'oci-eng', 'orm', 'orv-eng', 'pam-eng', 'pcm', 'pes-eng', 'pl', 'pl-en', 'pms-eng', 'pol-eng', 'por-eng', 'pt', 'ro', 'ron-eng', 'ru', 'run', 'rus-eng', 'sl', 'slk-eng', 'slv-eng', 'spa-eng', 'sna', 'som', 'sq', 'sqi-eng', 'srp-eng', 'sv', 'sw', 'swa', 'swe-eng', 'swg-eng', 'swh-eng', 'ta', 'tam-eng', 'tat-eng', 'te', 'tel-eng', 'tgl-eng', 'th', 'tha-eng', 'tir', 'tl', 'tr', 'tuk-eng', 'tur-eng', 'tzl-eng', 'uig-eng', 'ukr-eng', 'ur', 'urd-eng', 'uzb-eng', 'vi', 'vie-eng', 'war-eng', 'wuu-eng', 'xho', 'xho-eng', 'yid-eng', 'yor', 'yue-eng', 'zh', 'zh-CN', 'zh-TW', 'zh-en', 'zsm-eng']
27
+
28
+ # v_measures key is somehow present in voyage-2-law results and is a list
29
+ SKIP_KEYS = ["std", "evaluation_time", "main_score", "threshold", "v_measures"]
30
+
31
+ # Use "train" split instead
32
+ TRAIN_SPLIT = ["DanishPoliticalCommentsClassification"]
33
+ # Use "validation" split instead
34
+ VALIDATION_SPLIT = ["AFQMC", "Cmnli", "IFlyTek", "LEMBSummScreenFDRetrieval", "MSMARCO", "MSMARCO-PL", "MultilingualSentiment", "Ocnli", "TNews"]
35
+ # Use "dev" split instead
36
+ DEV_SPLIT = ["CmedqaRetrieval", "CovidRetrieval", "DuRetrieval", "EcomRetrieval", "MedicalRetrieval", "MMarcoReranking", "MMarcoRetrieval", "MSMARCO", "MSMARCO-PL", "T2Reranking", "T2Retrieval", "VideoRetrieval"]
37
+ # Use "test.full" split
38
+ TESTFULL_SPLIT = ["OpusparcusPC"]
39
+
40
+ TEST_AVG_SPLIT = {
41
+ "LEMBNeedleRetrieval": ["test_256", "test_512", "test_1024", "test_2048", "test_4096", "test_8192", "test_16384", "test_32768"],
42
+ "LEMBPasskeyRetrieval": ["test_256", "test_512", "test_1024", "test_2048", "test_4096", "test_8192", "test_16384", "test_32768"],
43
+ }
44
+
45
+ MODELS = [
46
+ "Baichuan-text-embedding",
47
+ "Cohere-embed-english-v3.0",
48
+ "Cohere-embed-multilingual-light-v3.0",
49
+ "Cohere-embed-multilingual-v3.0",
50
+ "DanskBERT",
51
+ "FollowIR-7B",
52
+ "GritLM-7B",
53
+ "LASER2",
54
+ "LLM2Vec-Llama-2-supervised",
55
+ "LLM2Vec-Llama-2-unsupervised",
56
+ "LLM2Vec-Meta-Llama-3-supervised",
57
+ "LLM2Vec-Meta-Llama-3-unsupervised",
58
+ "LLM2Vec-Mistral-supervised",
59
+ "LLM2Vec-Mistral-unsupervised",
60
+ "LLM2Vec-Sheared-Llama-supervised",
61
+ "LLM2Vec-Sheared-Llama-unsupervised",
62
+ "LaBSE",
63
+ "OpenSearch-text-hybrid",
64
+ "all-MiniLM-L12-v2",
65
+ "all-MiniLM-L6-v2",
66
+ "all-mpnet-base-v2",
67
+ "allenai-specter",
68
+ "bert-base-10lang-cased",
69
+ "bert-base-15lang-cased",
70
+ "bert-base-25lang-cased",
71
+ "bert-base-multilingual-cased",
72
+ "bert-base-multilingual-uncased",
73
+ "bert-base-swedish-cased",
74
+ "bert-base-uncased",
75
+ "bge-base-en-v1.5",
76
+ "bge-base-en",
77
+ "bge-base-zh",
78
+ "bge-base-zh-v1.5",
79
+ "bge-large-en",
80
+ "bge-large-en-v1.5",
81
+ "bge-large-zh",
82
+ "bge-large-zh-noinstruct",
83
+ "bge-large-zh-v1.5",
84
+ "bge-m3",
85
+ "bge-small-zh",
86
+ "bge-small-zh-v1.5",
87
+ "bm25",
88
+ "camembert-base",
89
+ "camembert-large",
90
+ "contriever-base-msmarco",
91
+ "cross-en-de-roberta-sentence-transformer",
92
+ "dfm-encoder-large-v1",
93
+ "dfm-sentence-encoder-large-1",
94
+ "distilbert-base-25lang-cased",
95
+ "distilbert-base-en-fr-cased",
96
+ "distilbert-base-en-fr-es-pt-it-cased",
97
+ "distilbert-base-fr-cased",
98
+ "distilbert-base-uncased",
99
+ "distiluse-base-multilingual-cased-v2",
100
+ "e5-base",
101
+ "e5-base-4k",
102
+ "e5-base-v2",
103
+ "e5-large",
104
+ "e5-large-v2",
105
+ "e5-mistral-7b-instruct",
106
+ "e5-small",
107
+ "e5-small-v2",
108
+ "electra-small-nordic",
109
+ "electra-small-swedish-cased-discriminator",
110
+ "elser-v2",
111
+ "embedder-100p",
112
+ "facebook-dpr-ctx_encoder-multiset-base",
113
+ "flan-t5-base",
114
+ "flan-t5-large",
115
+ "flaubert_base_cased",
116
+ "flaubert_base_uncased",
117
+ "flaubert_large_cased",
118
+ "gbert-base",
119
+ "gbert-large",
120
+ "gelectra-base",
121
+ "gelectra-large",
122
+ "glove.6B.300d",
123
+ "google-gecko-256.text-embedding-preview-0409",
124
+ "google-gecko.text-embedding-preview-0409",
125
+ "gottbert-base",
126
+ "gte-Qwen1.5-7B-instruct",
127
+ "gtr-t5-base",
128
+ "gtr-t5-large",
129
+ "gtr-t5-xl",
130
+ "gtr-t5-xxl",
131
+ "herbert-base-retrieval-v2",
132
+ "instructor-base",
133
+ "instructor-xl",
134
+ "jina-embeddings-v2-base-en",
135
+ "komninos",
136
+ "llama-2-7b-chat",
137
+ "luotuo-bert-medium",
138
+ "m3e-base",
139
+ "m3e-large",
140
+ "mistral-7b-instruct-v0.2",
141
+ "mistral-embed",
142
+ "monobert-large-msmarco",
143
+ "monot5-3b-msmarco-10k",
144
+ "monot5-base-msmarco-10k",
145
+ "msmarco-bert-co-condensor",
146
+ "multi-qa-MiniLM-L6-cos-v1",
147
+ "multilingual-e5-base",
148
+ "multilingual-e5-large",
149
+ "multilingual-e5-large-instruct",
150
+ "multilingual-e5-small",
151
+ "mxbai-embed-large-v1",
152
+ "nb-bert-base",
153
+ "nb-bert-large",
154
+ "nomic-embed-text-v1",
155
+ "nomic-embed-text-v1.5-128",
156
+ "nomic-embed-text-v1.5-256",
157
+ "nomic-embed-text-v1.5-512",
158
+ "nomic-embed-text-v1.5-64",
159
+ "norbert3-base",
160
+ "norbert3-large",
161
+ "paraphrase-multilingual-MiniLM-L12-v2",
162
+ "paraphrase-multilingual-mpnet-base-v2",
163
+ "rubert-tiny",
164
+ "rubert-tiny2",
165
+ "sbert_large_mt_nlu_ru",
166
+ "sbert_large_nlu_ru",
167
+ "sentence-bert-swedish-cased",
168
+ "sentence-camembert-base",
169
+ "sentence-camembert-large",
170
+ "sentence-croissant-llm-base",
171
+ "sentence-t5-base",
172
+ "sentence-t5-large",
173
+ "sentence-t5-xl",
174
+ "sentence-t5-xxl",
175
+ "sgpt-bloom-1b7-nli",
176
+ "sgpt-bloom-7b1-msmarco",
177
+ "silver-retriever-base-v1",
178
+ "st-polish-paraphrase-from-distilroberta",
179
+ "st-polish-paraphrase-from-mpnet",
180
+ "sup-simcse-bert-base-uncased",
181
+ "tart-dual-contriever-msmarco",
182
+ "tart-full-flan-t5-xl",
183
+ "text-embedding-3-large",
184
+ "text-embedding-3-large-256",
185
+ "text-embedding-3-small",
186
+ "text-embedding-ada-002",
187
+ "text-search-ada-001",
188
+ "text-search-ada-doc-001",
189
+ "text-search-babbage-001",
190
+ "text-search-curie-001",
191
+ "text-search-davinci-001",
192
+ "text-similarity-ada-001",
193
+ "text-similarity-babbage-001",
194
+ "text-similarity-curie-001",
195
+ "text-similarity-davinci-001",
196
+ "text2vec-base-chinese",
197
+ "text2vec-base-multilingual",
198
+ "text2vec-large-chinese",
199
+ "titan-embed-text-v1",
200
+ "udever-bloom-1b1",
201
+ "udever-bloom-560m",
202
+ "universal-sentence-encoder-multilingual-3",
203
+ "universal-sentence-encoder-multilingual-large-3",
204
+ "unsup-simcse-bert-base-uncased",
205
+ "use-cmlm-multilingual",
206
+ "voyage-2",
207
+ "voyage-code-2",
208
+ "voyage-large-2-instruct",
209
+ "voyage-law-2",
210
+ "voyage-lite-01-instruct",
211
+ "voyage-lite-02-instruct",
212
+ "voyage-multilingual-2",
213
+ "xlm-roberta-base",
214
+ "xlm-roberta-large",
215
+ ]
216
+
217
+
218
+ # Needs to be run whenever new files are added
219
+ def get_paths():
220
+ import collections, json, os
221
+ files = collections.defaultdict(list)
222
+ for model_dir in os.listdir("results"):
223
+ results_model_dir = os.path.join("results", model_dir)
224
+ if not os.path.isdir(results_model_dir):
225
+ print(f"Skipping {results_model_dir}")
226
+ continue
227
+ for revision_folder in os.listdir(results_model_dir):
228
+ if not os.path.isdir(os.path.join(results_model_dir, revision_folder)):
229
+ continue
230
+ for res_file in os.listdir(os.path.join(results_model_dir, revision_folder)):
231
+ if (res_file.endswith(".json")) and not(res_file.endswith("overall_results.json")):
232
+ results_model_file = os.path.join(results_model_dir, res_file)
233
+ files[model_dir].append(results_model_file)
234
+ with open("paths.json", "w") as f:
235
+ json.dump(files, f, indent=2)
236
+ return files
237
+
238
+
239
+ class MTEBResults(datasets.GeneratorBasedBuilder):
240
+ """MTEBResults"""
241
+
242
+ BUILDER_CONFIGS = [
243
+ datasets.BuilderConfig(
244
+ name=model,
245
+ description=f"{model} MTEB results",
246
+ version=VERSION,
247
+ )
248
+ for model in MODELS
249
+ ]
250
+
251
+ def _info(self):
252
+ return datasets.DatasetInfo(
253
+ description=_DESCRIPTION,
254
+ features=datasets.Features(
255
+ {
256
+ "mteb_dataset_name": datasets.Value("string"),
257
+ "eval_language": datasets.Value("string"),
258
+ "metric": datasets.Value("string"),
259
+ "score": datasets.Value("float"),
260
+ }
261
+ ),
262
+ supervised_keys=None,
263
+ citation=_CITATION,
264
+ )
265
+
266
+ def _split_generators(self, dl_manager):
267
+ path_file = dl_manager.download_and_extract(URL)
268
+ with open(path_file) as f:
269
+ files = json.load(f)
270
+
271
+ downloaded_files = dl_manager.download_and_extract(files[self.config.name])
272
+ return [
273
+ datasets.SplitGenerator(
274
+ name=datasets.Split.TEST,
275
+ gen_kwargs={'filepath': downloaded_files}
276
+ )
277
+ ]
278
+
279
+ def _generate_examples(self, filepath):
280
+ """This function returns the examples in the raw (text) form."""
281
+ logger.info(f"Generating examples from {filepath}")
282
+ out = []
283
+
284
+ for path in filepath:
285
+ with open(path, encoding="utf-8") as f:
286
+ res_dict = json.load(f)
287
+ # Naming changed from mteb_dataset_name to task_name
288
+ ds_name = res_dict.get("mteb_dataset_name", res_dict.get("task_name"))
289
+ # New MTEB format uses scores
290
+ res_dict = res_dict.get("scores", res_dict)
291
+
292
+ split = "test"
293
+ if (ds_name in TRAIN_SPLIT) and ("train" in res_dict):
294
+ split = "train"
295
+ elif (ds_name in VALIDATION_SPLIT) and ("validation" in res_dict):
296
+ split = "validation"
297
+ elif (ds_name in DEV_SPLIT) and ("dev" in res_dict):
298
+ split = "dev"
299
+ elif (ds_name in TESTFULL_SPLIT) and ("test.full" in res_dict):
300
+ split = "test.full"
301
+ elif (ds_name in TEST_AVG_SPLIT):
302
+ # Average splits
303
+ res_dict["test_avg"] = {}
304
+ for split in TEST_AVG_SPLIT[ds_name]:
305
+ # Old MTEB format
306
+ if isinstance(res_dict.get(split), dict):
307
+ for k, v in res_dict.get(split, {}).items():
308
+ v /= len(TEST_AVG_SPLIT[ds_name])
309
+ if k not in res_dict["test_avg"]:
310
+ res_dict["test_avg"][k] = v
311
+ else:
312
+ res_dict["test_avg"][k] += v
313
+ # New MTEB format
314
+ elif isinstance(res_dict.get(split), list):
315
+ assert len(res_dict[split]) == 1, "Only single-lists supported for now"
316
+ for k, v in res_dict[split][0].items():
317
+ if not isinstance(v, float): continue
318
+ v /= len(TEST_AVG_SPLIT[ds_name])
319
+ if k not in res_dict["test_avg"]:
320
+ res_dict["test_avg"][k] = v
321
+ else:
322
+ res_dict["test_avg"][k] += v
323
+ split = "test_avg"
324
+ elif "test" not in res_dict:
325
+ print(f"Skipping {ds_name} as split {split} not present.")
326
+ continue
327
+ res_dict = res_dict.get(split)
328
+
329
+ ### New MTEB format ###
330
+ if isinstance(res_dict, list):
331
+ for res in res_dict:
332
+ lang = res.get("languages", [""])
333
+ assert len(lang) == 1, "Only single-languages supported for now"
334
+ lang = lang[0].replace("eng-Latn", "")
335
+ for metric, score in res.items():
336
+ if metric in SKIP_KEYS: continue
337
+ out.append({
338
+ "mteb_dataset_name": ds_name,
339
+ "eval_language": lang,
340
+ "metric": metric,
341
+ "score": score * 100,
342
+ })
343
+
344
+ ### Old MTEB format ###
345
+ else:
346
+ is_multilingual = any(x in res_dict for x in EVAL_LANGS)
347
+ langs = res_dict.keys() if is_multilingual else ["en"]
348
+ for lang in langs:
349
+ if lang in SKIP_KEYS: continue
350
+ test_result_lang = res_dict.get(lang) if is_multilingual else res_dict
351
+ for metric, score in test_result_lang.items():
352
+ if not isinstance(score, dict):
353
+ score = {metric: score}
354
+ for sub_metric, sub_score in score.items():
355
+ if any(x in sub_metric for x in SKIP_KEYS): continue
356
+ if isinstance(sub_score, dict): continue
357
+ out.append({
358
+ "mteb_dataset_name": ds_name,
359
+ "eval_language": lang if is_multilingual else "",
360
+ "metric": f"{metric}_{sub_metric}" if metric != sub_metric else metric,
361
+ "score": sub_score * 100,
362
+ })
363
+ for idx, row in enumerate(sorted(out, key=lambda x: x["mteb_dataset_name"])):
364
+ yield idx, row
365
+
366
+
367
+ # NOTE: for generating the new paths
368
+ if __name__ == "__main__":
369
+ get_paths()
results/Baichuan-text-embedding/no_revision_available/AFQMC.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "AFQMC",
4
+ "mteb_version": "1.1.0",
5
+ "validation": {
6
+ "cos_sim": {
7
+ "pearson": 0.4829609272631085,
8
+ "spearman": 0.5080031098340034
9
+ },
10
+ "euclidean": {
11
+ "pearson": 0.48915888167383914,
12
+ "spearman": 0.508003310876931
13
+ },
14
+ "evaluation_time": 5.52,
15
+ "manhattan": {
16
+ "pearson": 0.4883913003371612,
17
+ "spearman": 0.507119124081868
18
+ }
19
+ }
20
+ }
results/Baichuan-text-embedding/no_revision_available/ATEC.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "ATEC",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "cos_sim": {
7
+ "pearson": 0.5108024865980523,
8
+ "spearman": 0.5322524599077678
9
+ },
10
+ "euclidean": {
11
+ "pearson": 0.5495649374475876,
12
+ "spearman": 0.5322525387159475
13
+ },
14
+ "evaluation_time": 23.5,
15
+ "manhattan": {
16
+ "pearson": 0.5490069145550858,
17
+ "spearman": 0.5318577305140235
18
+ }
19
+ }
20
+ }
results/Baichuan-text-embedding/no_revision_available/AmazonReviewsClassification.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "AmazonReviewsClassification",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 37.19,
7
+ "zh": {
8
+ "accuracy": 0.48301999999999995,
9
+ "accuracy_stderr": 0.015009983344427814,
10
+ "f1": 0.4358151996593711,
11
+ "f1_stderr": 0.021935952504451846,
12
+ "main_score": 0.48301999999999995
13
+ }
14
+ },
15
+ "validation": {
16
+ "evaluation_time": 44.97,
17
+ "zh": {
18
+ "accuracy": 0.47596,
19
+ "accuracy_stderr": 0.013457131938121147,
20
+ "f1": 0.4297519925445886,
21
+ "f1_stderr": 0.022271859166647427,
22
+ "main_score": 0.47596
23
+ }
24
+ }
25
+ }
results/Baichuan-text-embedding/no_revision_available/BQ.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "BQ",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "cos_sim": {
7
+ "pearson": 0.6382205659368678,
8
+ "spearman": 0.664875868977765
9
+ },
10
+ "euclidean": {
11
+ "pearson": 0.6469068710733963,
12
+ "spearman": 0.6648758689697865
13
+ },
14
+ "evaluation_time": 11.53,
15
+ "manhattan": {
16
+ "pearson": 0.6465254698872425,
17
+ "spearman": 0.664768204689053
18
+ }
19
+ }
20
+ }
results/Baichuan-text-embedding/no_revision_available/CLSClusteringP2P.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "CLSClusteringP2P",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 842.9,
7
+ "v_measure": 0.6036843826092417,
8
+ "v_measure_std": 0.011663855200498223
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/CLSClusteringS2S.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "CLSClusteringS2S",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 99.78,
7
+ "v_measure": 0.5108781005084964,
8
+ "v_measure_std": 0.01733772998995446
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/CMedQAv1.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "CMedQAv1",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 245.66,
7
+ "map": 0.8806336186199681,
8
+ "mrr": 0.9029142857142856
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/CMedQAv2.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "CMedQAv2",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 255.0,
7
+ "map": 0.8845694611153427,
8
+ "mrr": 0.9049976190476191
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/CmedqaRetrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 293.85,
5
+ "map_at_1": 0.27486,
6
+ "map_at_10": 0.40762,
7
+ "map_at_100": 0.42651,
8
+ "map_at_1000": 0.42753,
9
+ "map_at_3": 0.36423,
10
+ "map_at_5": 0.38844,
11
+ "mrr_at_1": 0.41935,
12
+ "mrr_at_10": 0.50255,
13
+ "mrr_at_100": 0.51194,
14
+ "mrr_at_1000": 0.51233,
15
+ "mrr_at_3": 0.47633,
16
+ "mrr_at_5": 0.49123,
17
+ "ndcg_at_1": 0.41935,
18
+ "ndcg_at_10": 0.47637,
19
+ "ndcg_at_100": 0.54772,
20
+ "ndcg_at_1000": 0.56477,
21
+ "ndcg_at_3": 0.42216,
22
+ "ndcg_at_5": 0.44463,
23
+ "precision_at_1": 0.41935,
24
+ "precision_at_10": 0.10463,
25
+ "precision_at_100": 0.01626,
26
+ "precision_at_1000": 0.00184,
27
+ "precision_at_3": 0.23731,
28
+ "precision_at_5": 0.17129,
29
+ "recall_at_1": 0.27486,
30
+ "recall_at_10": 0.58517,
31
+ "recall_at_100": 0.87758,
32
+ "recall_at_1000": 0.99031,
33
+ "recall_at_3": 0.42086,
34
+ "recall_at_5": 0.49157
35
+ },
36
+ "mteb_dataset_name": "CmedqaRetrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/Cmnli.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "Cmnli",
4
+ "mteb_version": "1.1.0",
5
+ "validation": {
6
+ "cos_sim": {
7
+ "accuracy": 0.7672880336740829,
8
+ "accuracy_threshold": 0.5732834339141846,
9
+ "ap": 0.853143557144231,
10
+ "f1": 0.7844515441959531,
11
+ "f1_threshold": 0.5262686014175415,
12
+ "precision": 0.7203207510267945,
13
+ "recall": 0.8611176057984569
14
+ },
15
+ "dot": {
16
+ "accuracy": 0.7672880336740829,
17
+ "accuracy_threshold": 0.5732833743095398,
18
+ "ap": 0.8530751882614949,
19
+ "f1": 0.7844515441959531,
20
+ "f1_threshold": 0.5262686610221863,
21
+ "precision": 0.7203207510267945,
22
+ "recall": 0.8611176057984569
23
+ },
24
+ "euclidean": {
25
+ "accuracy": 0.7672880336740829,
26
+ "accuracy_threshold": 0.9238144755363464,
27
+ "ap": 0.8531435417467815,
28
+ "f1": 0.7844515441959531,
29
+ "f1_threshold": 0.9733769297599792,
30
+ "precision": 0.7203207510267945,
31
+ "recall": 0.8611176057984569
32
+ },
33
+ "evaluation_time": 20.61,
34
+ "manhattan": {
35
+ "accuracy": 0.7683704149128082,
36
+ "accuracy_threshold": 23.47174644470215,
37
+ "ap": 0.853344222480869,
38
+ "f1": 0.784551963549577,
39
+ "f1_threshold": 24.62630844116211,
40
+ "precision": 0.7318356607974095,
41
+ "recall": 0.845452419920505
42
+ },
43
+ "max": {
44
+ "accuracy": 0.7683704149128082,
45
+ "ap": 0.853344222480869,
46
+ "f1": 0.784551963549577
47
+ }
48
+ }
49
+ }
results/Baichuan-text-embedding/no_revision_available/CovidRetrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 321.42,
5
+ "map_at_1": 0.75395,
6
+ "map_at_10": 0.83498,
7
+ "map_at_100": 0.83664,
8
+ "map_at_1000": 0.83666,
9
+ "map_at_3": 0.82183,
10
+ "map_at_5": 0.83043,
11
+ "mrr_at_1": 0.75553,
12
+ "mrr_at_10": 0.83455,
13
+ "mrr_at_100": 0.83613,
14
+ "mrr_at_1000": 0.83614,
15
+ "mrr_at_3": 0.82122,
16
+ "mrr_at_5": 0.8307,
17
+ "ndcg_at_1": 0.75553,
18
+ "ndcg_at_10": 0.86865,
19
+ "ndcg_at_100": 0.87511,
20
+ "ndcg_at_1000": 0.8754,
21
+ "ndcg_at_3": 0.84324,
22
+ "ndcg_at_5": 0.85864,
23
+ "precision_at_1": 0.75553,
24
+ "precision_at_10": 0.0981,
25
+ "precision_at_100": 0.01008,
26
+ "precision_at_1000": 0.00101,
27
+ "precision_at_3": 0.30313,
28
+ "precision_at_5": 0.18967,
29
+ "recall_at_1": 0.75395,
30
+ "recall_at_10": 0.97102,
31
+ "recall_at_100": 0.99789,
32
+ "recall_at_1000": 1.0,
33
+ "recall_at_3": 0.90516,
34
+ "recall_at_5": 0.94152
35
+ },
36
+ "mteb_dataset_name": "CovidRetrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/DuRetrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 331.03,
5
+ "map_at_1": 0.26579,
6
+ "map_at_10": 0.81391,
7
+ "map_at_100": 0.84156,
8
+ "map_at_1000": 0.84195,
9
+ "map_at_3": 0.56806,
10
+ "map_at_5": 0.71469,
11
+ "mrr_at_1": 0.9155,
12
+ "mrr_at_10": 0.94169,
13
+ "mrr_at_100": 0.94218,
14
+ "mrr_at_1000": 0.94221,
15
+ "mrr_at_3": 0.93908,
16
+ "mrr_at_5": 0.94081,
17
+ "ndcg_at_1": 0.9155,
18
+ "ndcg_at_10": 0.88427,
19
+ "ndcg_at_100": 0.90951,
20
+ "ndcg_at_1000": 0.91299,
21
+ "ndcg_at_3": 0.87826,
22
+ "ndcg_at_5": 0.86583,
23
+ "precision_at_1": 0.9155,
24
+ "precision_at_10": 0.4218,
25
+ "precision_at_100": 0.04808,
26
+ "precision_at_1000": 0.00489,
27
+ "precision_at_3": 0.7865,
28
+ "precision_at_5": 0.6624,
29
+ "recall_at_1": 0.26579,
30
+ "recall_at_10": 0.89204,
31
+ "recall_at_100": 0.97596,
32
+ "recall_at_1000": 0.99417,
33
+ "recall_at_3": 0.58946,
34
+ "recall_at_5": 0.7584
35
+ },
36
+ "mteb_dataset_name": "DuRetrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/EcomRetrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 52.09,
5
+ "map_at_1": 0.515,
6
+ "map_at_10": 0.6134,
7
+ "map_at_100": 0.61896,
8
+ "map_at_1000": 0.61911,
9
+ "map_at_3": 0.58683,
10
+ "map_at_5": 0.60308,
11
+ "mrr_at_1": 0.515,
12
+ "mrr_at_10": 0.6134,
13
+ "mrr_at_100": 0.61896,
14
+ "mrr_at_1000": 0.61911,
15
+ "mrr_at_3": 0.58683,
16
+ "mrr_at_5": 0.60308,
17
+ "ndcg_at_1": 0.515,
18
+ "ndcg_at_10": 0.66393,
19
+ "ndcg_at_100": 0.68905,
20
+ "ndcg_at_1000": 0.6928,
21
+ "ndcg_at_3": 0.61049,
22
+ "ndcg_at_5": 0.63975,
23
+ "precision_at_1": 0.515,
24
+ "precision_at_10": 0.0823,
25
+ "precision_at_100": 0.00936,
26
+ "precision_at_1000": 0.00097,
27
+ "precision_at_3": 0.22633,
28
+ "precision_at_5": 0.15,
29
+ "recall_at_1": 0.515,
30
+ "recall_at_10": 0.823,
31
+ "recall_at_100": 0.936,
32
+ "recall_at_1000": 0.965,
33
+ "recall_at_3": 0.679,
34
+ "recall_at_5": 0.75
35
+ },
36
+ "mteb_dataset_name": "EcomRetrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/IFlyTek.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "IFlyTek",
4
+ "mteb_version": "1.1.0",
5
+ "validation": {
6
+ "accuracy": 0.5075028857252789,
7
+ "accuracy_stderr": 0.003609400353846422,
8
+ "evaluation_time": 177.43,
9
+ "f1": 0.3640576261699708,
10
+ "f1_stderr": 0.007935818070069649,
11
+ "main_score": 0.5075028857252789
12
+ }
13
+ }
results/Baichuan-text-embedding/no_revision_available/JDReview.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "JDReview",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "accuracy": 0.876923076923077,
7
+ "accuracy_stderr": 0.015543951352501298,
8
+ "ap": 0.5771137523616632,
9
+ "ap_stderr": 0.027052492390223098,
10
+ "evaluation_time": 13.72,
11
+ "f1": 0.8269262506868096,
12
+ "f1_stderr": 0.01643864510374752,
13
+ "main_score": 0.876923076923077
14
+ }
15
+ }
results/Baichuan-text-embedding/no_revision_available/LCQMC.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "LCQMC",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "cos_sim": {
7
+ "pearson": 0.7329402204344446,
8
+ "spearman": 0.7659571213941639
9
+ },
10
+ "euclidean": {
11
+ "pearson": 0.7536085128119648,
12
+ "spearman": 0.7659571213847517
13
+ },
14
+ "evaluation_time": 10.5,
15
+ "manhattan": {
16
+ "pearson": 0.7532685583893368,
17
+ "spearman": 0.7656404415349796
18
+ }
19
+ }
20
+ }
results/Baichuan-text-embedding/no_revision_available/MMarcoReranking.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "MMarcoReranking",
4
+ "mteb_version": "1.1.0",
5
+ "dev": {
6
+ "evaluation_time": 353.06,
7
+ "map": 0.34300784663135503,
8
+ "mrr": 0.3397539682539683
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/MMarcoRetrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 164.09,
5
+ "map_at_1": 0.67579,
6
+ "map_at_10": 0.76596,
7
+ "map_at_100": 0.76885,
8
+ "map_at_1000": 0.76895,
9
+ "map_at_3": 0.74784,
10
+ "map_at_5": 0.75943,
11
+ "mrr_at_1": 0.69814,
12
+ "mrr_at_10": 0.77122,
13
+ "mrr_at_100": 0.77379,
14
+ "mrr_at_1000": 0.77388,
15
+ "mrr_at_3": 0.75583,
16
+ "mrr_at_5": 0.76558,
17
+ "ndcg_at_1": 0.69814,
18
+ "ndcg_at_10": 0.80173,
19
+ "ndcg_at_100": 0.81457,
20
+ "ndcg_at_1000": 0.81738,
21
+ "ndcg_at_3": 0.76788,
22
+ "ndcg_at_5": 0.78716,
23
+ "precision_at_1": 0.69814,
24
+ "precision_at_10": 0.09642,
25
+ "precision_at_100": 0.01027,
26
+ "precision_at_1000": 0.00105,
27
+ "precision_at_3": 0.28806,
28
+ "precision_at_5": 0.18318,
29
+ "recall_at_1": 0.67579,
30
+ "recall_at_10": 0.90635,
31
+ "recall_at_100": 0.96408,
32
+ "recall_at_1000": 0.98619,
33
+ "recall_at_3": 0.81756,
34
+ "recall_at_5": 0.86314
35
+ },
36
+ "mteb_dataset_name": "MMarcoRetrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/MassiveIntentClassification.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "MassiveIntentClassification",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 21.27,
7
+ "zh-CN": {
8
+ "accuracy": 0.749058507061197,
9
+ "accuracy_stderr": 0.011884386350506135,
10
+ "f1": 0.7161970536275908,
11
+ "f1_stderr": 0.016191767473599423,
12
+ "main_score": 0.749058507061197
13
+ }
14
+ },
15
+ "validation": {
16
+ "evaluation_time": 19.95,
17
+ "zh-CN": {
18
+ "accuracy": 0.7576487948844074,
19
+ "accuracy_stderr": 0.01332642347320569,
20
+ "f1": 0.7118884143366013,
21
+ "f1_stderr": 0.015415315352111523,
22
+ "main_score": 0.7576487948844074
23
+ }
24
+ }
25
+ }
results/Baichuan-text-embedding/no_revision_available/MassiveScenarioClassification.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "MassiveScenarioClassification",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 13.92,
7
+ "zh-CN": {
8
+ "accuracy": 0.8128110289172831,
9
+ "accuracy_stderr": 0.010581970443298772,
10
+ "f1": 0.7981971014179347,
11
+ "f1_stderr": 0.013781985332394563,
12
+ "main_score": 0.8128110289172831
13
+ }
14
+ },
15
+ "validation": {
16
+ "evaluation_time": 16.58,
17
+ "zh-CN": {
18
+ "accuracy": 0.8044269552385638,
19
+ "accuracy_stderr": 0.01384668511219742,
20
+ "f1": 0.7893222744509287,
21
+ "f1_stderr": 0.01724155206070888,
22
+ "main_score": 0.8044269552385638
23
+ }
24
+ }
25
+ }
results/Baichuan-text-embedding/no_revision_available/MedicalRetrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 142.23,
5
+ "map_at_1": 0.523,
6
+ "map_at_10": 0.583,
7
+ "map_at_100": 0.5883,
8
+ "map_at_1000": 0.58884,
9
+ "map_at_3": 0.572,
10
+ "map_at_5": 0.57715,
11
+ "mrr_at_1": 0.524,
12
+ "mrr_at_10": 0.58355,
13
+ "mrr_at_100": 0.58885,
14
+ "mrr_at_1000": 0.58939,
15
+ "mrr_at_3": 0.5725,
16
+ "mrr_at_5": 0.5777,
17
+ "ndcg_at_1": 0.523,
18
+ "ndcg_at_10": 0.61098,
19
+ "ndcg_at_100": 0.63829,
20
+ "ndcg_at_1000": 0.65398,
21
+ "ndcg_at_3": 0.58721,
22
+ "ndcg_at_5": 0.59659,
23
+ "precision_at_1": 0.523,
24
+ "precision_at_10": 0.0699,
25
+ "precision_at_100": 0.0083,
26
+ "precision_at_1000": 0.00096,
27
+ "precision_at_3": 0.21033,
28
+ "precision_at_5": 0.1308,
29
+ "recall_at_1": 0.523,
30
+ "recall_at_10": 0.699,
31
+ "recall_at_100": 0.83,
32
+ "recall_at_1000": 0.956,
33
+ "recall_at_3": 0.631,
34
+ "recall_at_5": 0.654
35
+ },
36
+ "mteb_dataset_name": "MedicalRetrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/MultilingualSentiment.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "MultilingualSentiment",
4
+ "mteb_version": "1.1.0",
5
+ "validation": {
6
+ "accuracy": 0.7683333333333333,
7
+ "accuracy_stderr": 0.002808716591058785,
8
+ "evaluation_time": 35.42,
9
+ "f1": 0.7634474528405593,
10
+ "f1_stderr": 0.004128334463667356,
11
+ "main_score": 0.7683333333333333
12
+ }
13
+ }
results/Baichuan-text-embedding/no_revision_available/Ocnli.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "Ocnli",
4
+ "mteb_version": "1.1.0",
5
+ "validation": {
6
+ "cos_sim": {
7
+ "accuracy": 0.7552788305360043,
8
+ "accuracy_threshold": 0.5918081402778625,
9
+ "ap": 0.7933459144431134,
10
+ "f1": 0.7710033605376861,
11
+ "f1_threshold": 0.5680263042449951,
12
+ "precision": 0.7068661971830986,
13
+ "recall": 0.8479408658922915
14
+ },
15
+ "dot": {
16
+ "accuracy": 0.7552788305360043,
17
+ "accuracy_threshold": 0.5918081402778625,
18
+ "ap": 0.7933459144431134,
19
+ "f1": 0.7710033605376861,
20
+ "f1_threshold": 0.5680263042449951,
21
+ "precision": 0.7068661971830986,
22
+ "recall": 0.8479408658922915
23
+ },
24
+ "euclidean": {
25
+ "accuracy": 0.7552788305360043,
26
+ "accuracy_threshold": 0.9035395383834839,
27
+ "ap": 0.7933459144431134,
28
+ "f1": 0.7710033605376861,
29
+ "f1_threshold": 0.9294876456260681,
30
+ "precision": 0.7068661971830986,
31
+ "recall": 0.8479408658922915
32
+ },
33
+ "evaluation_time": 2.86,
34
+ "manhattan": {
35
+ "accuracy": 0.7531131564699513,
36
+ "accuracy_threshold": 22.5145263671875,
37
+ "ap": 0.7936757512903327,
38
+ "f1": 0.7693023255813953,
39
+ "f1_threshold": 24.065532684326172,
40
+ "precision": 0.6874480465502909,
41
+ "recall": 0.8732840549102429
42
+ },
43
+ "max": {
44
+ "accuracy": 0.7552788305360043,
45
+ "ap": 0.7936757512903327,
46
+ "f1": 0.7710033605376861
47
+ }
48
+ }
49
+ }
results/Baichuan-text-embedding/no_revision_available/OnlineShopping.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "OnlineShopping",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "accuracy": 0.9441999999999998,
7
+ "accuracy_stderr": 0.0028913664589601948,
8
+ "ap": 0.9253177971136661,
9
+ "ap_stderr": 0.004435733302762033,
10
+ "evaluation_time": 15.57,
11
+ "f1": 0.9440757676006394,
12
+ "f1_stderr": 0.0028698839591287813,
13
+ "main_score": 0.9441999999999998
14
+ }
15
+ }
results/Baichuan-text-embedding/no_revision_available/PAWSX.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "PAWSX",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "cos_sim": {
7
+ "pearson": 0.4440187963946336,
8
+ "spearman": 0.47560337376825085
9
+ },
10
+ "euclidean": {
11
+ "pearson": 0.47582142557969154,
12
+ "spearman": 0.47560337376825085
13
+ },
14
+ "evaluation_time": 2.49,
15
+ "manhattan": {
16
+ "pearson": 0.4752151199528625,
17
+ "spearman": 0.47509647784531567
18
+ }
19
+ }
20
+ }
results/Baichuan-text-embedding/no_revision_available/QBQTC.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "QBQTC",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "cos_sim": {
7
+ "pearson": 0.3862883070440486,
8
+ "spearman": 0.39955352213272627
9
+ },
10
+ "euclidean": {
11
+ "pearson": 0.3786832501667679,
12
+ "spearman": 0.3995535221519048
13
+ },
14
+ "evaluation_time": 5.97,
15
+ "manhattan": {
16
+ "pearson": 0.3783822070118956,
17
+ "spearman": 0.3990713604918495
18
+ }
19
+ }
20
+ }
results/Baichuan-text-embedding/no_revision_available/STS22.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": "6d1ba47164174a496b7fa5d3569dae26a6813b80",
3
+ "mteb_dataset_name": "STS22",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 7.61,
7
+ "zh": {
8
+ "cos_sim": {
9
+ "pearson": 0.6082900117033369,
10
+ "spearman": 0.657843998948187
11
+ },
12
+ "euclidean": {
13
+ "pearson": 0.6276823760827268,
14
+ "spearman": 0.657843998948187
15
+ },
16
+ "manhattan": {
17
+ "pearson": 0.6312607312720664,
18
+ "spearman": 0.6575028602582914
19
+ }
20
+ }
21
+ }
22
+ }
results/Baichuan-text-embedding/no_revision_available/STSB.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "STSB",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "cos_sim": {
7
+ "pearson": 0.8067336246251158,
8
+ "spearman": 0.8014220624121933
9
+ },
10
+ "euclidean": {
11
+ "pearson": 0.7903263059130334,
12
+ "spearman": 0.8014220624121933
13
+ },
14
+ "evaluation_time": 2.16,
15
+ "manhattan": {
16
+ "pearson": 0.789560236524975,
17
+ "spearman": 0.7997430607737568
18
+ }
19
+ }
20
+ }
results/Baichuan-text-embedding/no_revision_available/T2Reranking.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "T2Reranking",
4
+ "mteb_version": "1.1.0",
5
+ "dev": {
6
+ "evaluation_time": 767.43,
7
+ "map": 0.6784944108203279,
8
+ "mrr": 0.7791862795671192
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/T2Retrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 626.32,
5
+ "map_at_1": 0.25021,
6
+ "map_at_10": 0.71544,
7
+ "map_at_100": 0.75456,
8
+ "map_at_1000": 0.75537,
9
+ "map_at_3": 0.49671,
10
+ "map_at_5": 0.61299,
11
+ "mrr_at_1": 0.85534,
12
+ "mrr_at_10": 0.89019,
13
+ "mrr_at_100": 0.89152,
14
+ "mrr_at_1000": 0.89158,
15
+ "mrr_at_3": 0.88413,
16
+ "mrr_at_5": 0.8879,
17
+ "ndcg_at_1": 0.85534,
18
+ "ndcg_at_10": 0.8011,
19
+ "ndcg_at_100": 0.84517,
20
+ "ndcg_at_1000": 0.85311,
21
+ "ndcg_at_3": 0.81212,
22
+ "ndcg_at_5": 0.79782,
23
+ "precision_at_1": 0.85534,
24
+ "precision_at_10": 0.40517,
25
+ "precision_at_100": 0.04961,
26
+ "precision_at_1000": 0.00514,
27
+ "precision_at_3": 0.71484,
28
+ "precision_at_5": 0.60081,
29
+ "recall_at_1": 0.25021,
30
+ "recall_at_10": 0.79721,
31
+ "recall_at_100": 0.93976,
32
+ "recall_at_1000": 0.98037,
33
+ "recall_at_3": 0.51869,
34
+ "recall_at_5": 0.65616
35
+ },
36
+ "mteb_dataset_name": "T2Retrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/TNews.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "TNews",
4
+ "mteb_version": "1.1.0",
5
+ "validation": {
6
+ "accuracy": 0.5261899999999999,
7
+ "accuracy_stderr": 0.005332813516334514,
8
+ "evaluation_time": 80.03,
9
+ "f1": 0.5088699308193914,
10
+ "f1_stderr": 0.005808330018430932,
11
+ "main_score": 0.5261899999999999
12
+ }
13
+ }
results/Baichuan-text-embedding/no_revision_available/ThuNewsClusteringP2P.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "ThuNewsClusteringP2P",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 684.88,
7
+ "v_measure": 0.582283860392516,
8
+ "v_measure_std": 0.02478622759991042
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/ThuNewsClusteringS2S.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "ThuNewsClusteringS2S",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "evaluation_time": 71.89,
7
+ "v_measure": 0.5782974132365385,
8
+ "v_measure_std": 0.013022585576314076
9
+ }
10
+ }
results/Baichuan-text-embedding/no_revision_available/VideoRetrieval.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "dev": {
4
+ "evaluation_time": 56.91,
5
+ "map_at_1": 0.606,
6
+ "map_at_10": 0.70041,
7
+ "map_at_100": 0.70481,
8
+ "map_at_1000": 0.70494,
9
+ "map_at_3": 0.682,
10
+ "map_at_5": 0.6925,
11
+ "mrr_at_1": 0.606,
12
+ "mrr_at_10": 0.70041,
13
+ "mrr_at_100": 0.70481,
14
+ "mrr_at_1000": 0.70494,
15
+ "mrr_at_3": 0.682,
16
+ "mrr_at_5": 0.6925,
17
+ "ndcg_at_1": 0.606,
18
+ "ndcg_at_10": 0.74282,
19
+ "ndcg_at_100": 0.7629,
20
+ "ndcg_at_1000": 0.76595,
21
+ "ndcg_at_3": 0.70476,
22
+ "ndcg_at_5": 0.72369,
23
+ "precision_at_1": 0.606,
24
+ "precision_at_10": 0.0875,
25
+ "precision_at_100": 0.00966,
26
+ "precision_at_1000": 0.00099,
27
+ "precision_at_3": 0.25667,
28
+ "precision_at_5": 0.1632,
29
+ "recall_at_1": 0.606,
30
+ "recall_at_10": 0.875,
31
+ "recall_at_100": 0.966,
32
+ "recall_at_1000": 0.989,
33
+ "recall_at_3": 0.77,
34
+ "recall_at_5": 0.816
35
+ },
36
+ "mteb_dataset_name": "VideoRetrieval",
37
+ "mteb_version": "1.1.0"
38
+ }
results/Baichuan-text-embedding/no_revision_available/Waimai.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": null,
3
+ "mteb_dataset_name": "Waimai",
4
+ "mteb_version": "1.1.0",
5
+ "test": {
6
+ "accuracy": 0.8877,
7
+ "accuracy_stderr": 0.005478138369920941,
8
+ "ap": 0.7410049925615304,
9
+ "ap_stderr": 0.01151741307800683,
10
+ "evaluation_time": 13.59,
11
+ "f1": 0.8718799117513892,
12
+ "f1_stderr": 0.005127107574743738,
13
+ "main_score": 0.8877
14
+ }
15
+ }
results/Cohere-embed-english-v3.0/no_revision_available/AILACasedocs.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.31543}, "mteb_dataset_name": "AILACasedocs"}
results/Cohere-embed-english-v3.0/no_revision_available/AILAStatutes.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.27152}, "mteb_dataset_name": "AILAStatutes"}
results/Cohere-embed-english-v3.0/no_revision_available/Core17InstructionRetrieval.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dataset_revision": "e39ff896cf3efbbdeeb950e6bd7c79f266995b07",
3
+ "mteb_dataset_name": "Core17InstructionRetrieval",
4
+ "mteb_version": "1.7.32",
5
+ "test": {
6
+ "evaluation_time": 746.94,
7
+ "p-MRR": 0.028043926455402175
8
+ }
9
+ }
results/Cohere-embed-english-v3.0/no_revision_available/GerDaLIRSmall.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.06047}, "mteb_dataset_name": "GerDaLIRSmall"}
results/Cohere-embed-english-v3.0/no_revision_available/LeCaRDv2.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.21017}, "mteb_dataset_name": "LeCaRDv2"}
results/Cohere-embed-english-v3.0/no_revision_available/LegalBenchConsumerContractsQA.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.7712}, "mteb_dataset_name": "LegalBenchConsumerContractsQA"}
results/Cohere-embed-english-v3.0/no_revision_available/LegalBenchCorporateLobbying.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.93681}, "mteb_dataset_name": "LegalBenchCorporateLobbying"}
results/Cohere-embed-english-v3.0/no_revision_available/LegalQuAD.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.26075}, "mteb_dataset_name": "LegalQuAD"}
results/Cohere-embed-english-v3.0/no_revision_available/LegalSummarization.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"test": {"ndcg_at_10": 0.61697}, "mteb_dataset_name": "LegalSummarization"}