Commit
•
9084414
1
Parent(s):
ca5798f
Update README.md
Browse files
README.md
CHANGED
|
@@ -38,7 +38,8 @@ language:
|
|
| 38 |
- 中英双语，以及中英跨语种能力(Bilingual and Crosslingual capability in English and Chinese)；
|
| 39 |
- RAG优化，适配更多真实业务场景(RAG adaptation for more domains, including Education, Law, Finance, Medical, Literature, FAQ, Textbook, Wikipedia, etc.)；
|
| 40 |
- 方便集成进langchain和llamaindex(Easy integrations for langchain and llamaindex in <a href="https://github.com/netease-youdao/BCEmbedding">BCEmbedding</a>)。
|
| 41 |
-
- `EmbeddingModel`不需要“精心设计”instruction，尽可能召回有用片段。
|
|
|
|
| 42 |
|
| 43 |
## News:
|
| 44 |
- `BCEmbedding`技术博客（ **Technical Blog** ）: [为RAG而生-BCEmbedding技术报告](https://zhuanlan.zhihu.com/p/681370855)
|
|
|
|
| 38 |
- 中英双语，以及中英跨语种能力(Bilingual and Crosslingual capability in English and Chinese)；
|
| 39 |
- RAG优化，适配更多真实业务场景(RAG adaptation for more domains, including Education, Law, Finance, Medical, Literature, FAQ, Textbook, Wikipedia, etc.)；
|
| 40 |
- 方便集成进langchain和llamaindex(Easy integrations for langchain and llamaindex in <a href="https://github.com/netease-youdao/BCEmbedding">BCEmbedding</a>)。
|
| 41 |
+
- `EmbeddingModel`不需要“精心设计”instruction，尽可能召回有用片段。 (No need for "instruction")
|
| 42 |
+
- **最佳实践（Best practice）** ：embedding召回top50-100片段，reranker对这50-100片段精排，最后取top5-10片段。（1. Get top 50-100 passages with [bce-embedding-base_v1](https://huggingface.co/maidalun1020/bce-embedding-base_v1) for "`recall`"； 2. Rerank passages with [bce-reranker-base_v1](https://huggingface.co/maidalun1020/bce-reranker-base_v1) and get top 5-10 for "`precision`" finally. ）
|
| 43 |
|
| 44 |
## News:
|
| 45 |
- `BCEmbedding`技术博客（ **Technical Blog** ）: [为RAG而生-BCEmbedding技术报告](https://zhuanlan.zhihu.com/p/681370855)
|