update readme
Browse files
README.md
CHANGED
|
@@ -1055,4 +1055,19 @@ model-index:
|
|
| 1055 |
value: 88.2026184757175
|
| 1056 |
---
|
| 1057 |
|
| 1058 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1055 |
value: 88.2026184757175
|
| 1056 |
---
|
| 1057 |
|
| 1058 |
+
**新闻 | News**
|
| 1059 |
+
|
| 1060 |
+
**[2024-04-22]**
|
| 1061 |
+
|
| 1062 |
+
piccolo-large-zh-v2 目前在C-MTEB榜单取得第一名，领先上一名BERT模型约1.9个点。
|
| 1063 |
+
piccolo-large-zh-v2 currently ranks first on the C-MTEB list, leading the previous BERT model by about 1.9 points.
|
| 1064 |
+
|
| 1065 |
+
## piccolo-large-zh-v2
|
| 1066 |
+
|
| 1067 |
+
piccolo-large-zh-v2 是一个通用embedding模型(中文), 由来自商汤科技的通用模型组完成训练，此次piccolo升级旨在更多地关注通用的下游finetune方式。
|
| 1068 |
+
目前该模型暂时需要通过API来进行访问: https://platform.sensenova.cn/doc?path=/chat/Embeddings/Embeddings.md
|
| 1069 |
+
我们将在近期更新我们的技术报告，同时详细技术细节也将在商汤4.23技术交流日披露。
|
| 1070 |
+
|
| 1071 |
+
piccolo-large-zh-v2 is an embedding model (Chinese), trained by the general model group from SenseTime Reserach. This piccolo upgrade aims to pay more attention to the general downstream finetune method.
|
| 1072 |
+
Currently, the model needs to be accessed through API: https://platform.sensenova.cn/doc?path=/chat/Embeddings/Embeddings.md
|
| 1073 |
+
We will update our technical report in the near future, and detailed technical details will also be disclosed on SenseTime 4.23 Tech Day.
|