Update README.md

README.md

@@ -38,7 +38,7 @@ language:
 - 中英日韩四个语种，以及中英日韩四个语种的跨语种能力(Multilingual and Crosslingual capability in English, Chinese, Japanese and Korean)；
 - 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.)；
 - <a href="https://github.com/netease-youdao/BCEmbedding">BCEmbedding</a>适配长文本做rerank(Handle long passages reranking more than 512 limit in <a href="https://github.com/netease-youdao/BCEmbedding">BCEmbedding</a>)；
-- RerankerModel可以提供 **“平滑”的“绝对”相关性分数**，**“平滑”对排序友好**，**“绝对”分数用于过滤低质量passage**。（RerankerModel provides **"smooth" (for reranking) and "meaningful" (for filtering bad passages) similarity score**, which help you figure out how relavent the query and passages are!）
+- RerankerModel可以提供 **“平滑”的“绝对”相关性分数**，**“平滑”对排序友好**，**“绝对”分数用于过滤低质量passage**，低质量passage过滤阈值推荐0.35或0.4。（RerankerModel provides **"smooth" (for reranking) and "meaningful" (for filtering bad passages with a threshold of 0.35 or 0.4) similarity score**, which help you figure out how relavent the query and passages are!）
 - **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. ）
 
 ## News: