jupyterjazz commited on
Commit
345e984
1 Parent(s): 2ae59f1

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +53 -1
README.md CHANGED
@@ -264,7 +264,59 @@ embeddings = model.encode(['How is the weather today?'], task_type='retrieval.qu
264
 
265
  ## Performance
266
 
267
- TODO UPDATE THIS
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
268
 
269
  ## Contact
270
 
 
264
 
265
  ## Performance
266
 
267
+ ### English MTEB
268
+ | Model | Average | CF | CL | PC | RR | RT | STS | SM |
269
+ |:------------------------------:|:-------:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|
270
+ | jina-embeddings-v2-en | 58.12 | 68.82| 40.08| 84.44| 55.09| 45.64| 80.00| 30.56|
271
+ | jina-embeddings-v3 | **65.60** | **82.58**| 45.27| 84.01| 58.13| 53.87| **85.8** | 30.98|
272
+ | text-embedding-3-large | 62.03 | 75.45| 49.01| 84.22| 59.16| 55.44| 81.04| 29.92|
273
+ | multilingual-e5-large-instruct | 64.41 | 77.56| 47.1 | 86.19| 58.58| 52.47| 84.78| 30.39|
274
+ | Cohere-embed-multilingual-v3.0 | 60.08 | 64.01| 46.6 | 86.15| 57.86| 53.84| 83.15| 30.99|
275
+
276
+ ### Multilingual MTEB
277
+
278
+ | Model | Average | CF | CL | PC | RR | RT | STS | SM |
279
+ |:------------------------------:|:-------:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|
280
+ | jina-embeddings-v2* | 60.54 | 65.69| 39.36| **82.95**| 66.57| 58.24| 66.6 | - |
281
+ | jina-embeddings-v3 | **64.44** | **71.46**| 46.71| 76.91| 63.98| 57.98| **69.83**| - |
282
+ | multilingual-e5-large | 59.58 | 65.22| 42.12| 76.95| 63.4 | 52.37| 64.65| - |
283
+ | multilingual-e5-large-instruct | 64.25 | 67.45| **52.12**| 77.79| **69.02**| **58.38**| 68.77| - |
284
+
285
+ **Notes:**
286
+ - CF: Classification Accuracy
287
+ - CL: Clustering V measure
288
+ - PC: Pair Classification Average Precision
289
+ - RR: Reranking MAP
290
+ - RT: Retrieval nDCG@10
291
+ - STS: Sentence Similarity Spearman Correlation
292
+ - SM: Summarization Spearman Correlation
293
+ - `*`: Jina Embeddings v2 bilingual model suite, only evaluated on Chinese, Spanish and German tasks.
294
+
295
+ ### Long Context Tasks (LongEmbed)
296
+
297
+ | Model | Average | NarrativeQA | Needle | Passkey | QMSum | SummScreen | WikiQA |
298
+ |:--------------------:|:-------:|:-----------:|:------:|:-------:|:-----:|:----------:|:------:|
299
+ | jina-embeddings-v3* | **70.39** | 33.32 | **84.00** | **100.00** | **39.75** | 92.78 | 72.46 |
300
+ | jina-embeddings-v2 | 58.12 | 37.89 | 54.25 | 50.25 | 38.87 | 93.48 | 73.99 |
301
+ | text-embedding-3-large | 51.3 | 44.09 | 29.25 | 63.00 | 32.49 | 84.80 | 54.16 |
302
+ | baai-bge-m3 | 56.56 | **45.76** | 40.25 | 46.00 | 35.54 | **94.09** | **77.73** |
303
+
304
+ **Notes:**
305
+ - `*`: text-matching adapter
306
+
307
+
308
+ #### Matryoshka Embeddings
309
+
310
+ | MRL Dimension | Retrieval | STS |
311
+ |:-------------:|:---------:|:----:|
312
+ | 32 | 52.54 | 76.35|
313
+ | 64 | 58.54 | 77.03|
314
+ | 128 | 61.64 | 77.43|
315
+ | 256 | 62.72 | 77.56|
316
+ | 512 | 63.16 | 77.59|
317
+ | 768 | 63.30 | 77.59|
318
+ | 1024 | 63.35 | 77.58|
319
+
320
 
321
  ## Contact
322