Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,46 @@
|
|
1 |
---
|
2 |
license: mit
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
+
language:
|
4 |
+
- zh
|
5 |
+
pipeline_tag: sentence-similarity
|
6 |
---
|
7 |
+
|
8 |
+
|
9 |
+
# PromCSE(sup)
|
10 |
+
|
11 |
+
|
12 |
+
|
13 |
+
|
14 |
+
|
15 |
+
## Data List
|
16 |
+
The following datasets are all in Chinese.
|
17 |
+
| Data | size(train) | size(valid) | size(test) |
|
18 |
+
|:----------------------:|:----------:|:----------:|:----------:|
|
19 |
+
| [STS-B](https://link.zhihu.com/?target=https%3A//pan.baidu.com/s/10yfKfTtcmLQ70-jzHIln1A%3Fpwd%3Dgf8y) | 5231| 1458| 1361|
|
20 |
+
| [ATEC](https://link.zhihu.com/?target=https%3A//pan.baidu.com/s/1gmnyz9emqOXwaHhSM9CCUA%3Fpwd%3Db17c) | 62477| 20000| 20000|
|
21 |
+
| [BQ](https://link.zhihu.com/?target=https%3A//pan.baidu.com/s/1M-e01yyy5NacVPrph9fbaQ%3Fpwd%3Dtis9) | 100000| 10000| 10000|
|
22 |
+
| [LCQMC](https://pan.baidu.com/s/16DfE7fHrCkk4e8a2j3SYUg?pwd=bc8w ) | 238766| 8802| 12500|
|
23 |
+
| [PAWSX](https://link.zhihu.com/?target=https%3A//pan.baidu.com/s/1ox0tJY3ZNbevHDeAqDBOPQ%3Fpwd%3Dmgjn) | 49401| 2000| 2000|
|
24 |
+
| [SNLI](https://link.zhihu.com/?target=https%3A//pan.baidu.com/s/1NOgA7JwWghiauwGAUvcm7w%3Fpwd%3Ds75v) | 146828| 2699| 2618|
|
25 |
+
| [MNLI](https://link.zhihu.com/?target=https%3A//pan.baidu.com/s/1xjZKtWk3MAbJ6HX4pvXJ-A%3Fpwd%3D2kte) | 122547| 2932| 2397|
|
26 |
+
|
27 |
+
|
28 |
+
|
29 |
+
|
30 |
+
|
31 |
+
|
32 |
+
## Model List
|
33 |
+
The evaluation dataset is in Chinese, and we used the same language model **RoBERTa Large** on different methods. In addition, considering that the test set of some datasets is small, which may lead to a large deviation in evaluation accuracy, the evaluation data here uses train, valid and test at the same time, and the final evaluation result adopts the **weighted average (w-avg)** method.
|
34 |
+
|
35 |
+
| Model | STS-B(w-avg) | ATEC | BQ | LCQMC | PAWSX | Avg. |
|
36 |
+
|:-----------------------:|:------------:|:-----------:|:----------|:----------|:----------:|:----------:|
|
37 |
+
| [BAAI/bge-large-zh](https://huggingface.co/BAAI/bge-large-zh) | 78.61| -| -| -| -| -|
|
38 |
+
| [BAAI/bge-large-zh-v1.5](https://huggingface.co/BAAI/bge-large-zh-v1.5) | 79.07| -| -| -| -| -|
|
39 |
+
| [hellonlp/simcse-large-zh](https://huggingface.co/hellonlp/simcse-roberta-large-zh) | 81.32| -| -| -| -| -|
|
40 |
+
| [hellonlp/promcse-large-zh](https://huggingface.co/hellonlp/promcse-bert-large-zh) | xx| -| -| -| -| -|
|
41 |
+
|
42 |
+
|
43 |
+
|
44 |
+
|
45 |
+
|
46 |
+
|