Update README.md

README.md

@@ -8,33 +8,33 @@ tags:
 - sentence-transformers
 library_name: transformers
 ---
-## UltraRAG-Reranker
+## MiniCPM-Reranker-Light
 
-**UltraRAG-Reranker** 是面壁智能与清华大学自然语言处理实验室（THUNLP）、东北大学信息检索小组（NEUIR）共同开发的中英双语言文本重排序模型，有如下特点：
+**MiniCPM-Reranker-Light** 是面壁智能与清华大学自然语言处理实验室（THUNLP）、东北大学信息检索小组（NEUIR）共同开发的中英双语言文本重排序模型，有如下特点：
 - 出色的中文、英文重排序能力。
 - 出色的中英跨语言重排序能力。
 - 支持长文本（最长8192token）。
 
-UltraRAG-Reranker 基于 [MiniCPM-1B-sft-bf16](https://huggingface.co/openbmb/MiniCPM-1B-sft-bf16) 训练，结构上采取双向注意力。采取多阶段训练方式，共使用包括开源数据、机造数据、闭源数据在内的约 500 万条训练数据。
+MiniCPM-Reranker-Light 基于 [MiniCPM-1B-sft-bf16](https://huggingface.co/openbmb/MiniCPM-1B-sft-bf16) 训练，结构上采取双向注意力。采取多阶段训练方式，共使用包括开源数据、机造数据、闭源数据在内的约 500 万条训练数据。
 
 欢迎关注 UltraRAG 系列：
 
-- 检索模型：[UltraRAG-Embedding](https://huggingface.co/openbmb/UltraRAG-Embedding)
-- 重排模型：[UltraRAG-Reranker](https://huggingface.co/openbmb/UltraRAG-Reranker)
+- 检索模型：[MiniCPM-Embedding-Light](https://huggingface.co/openbmb/MiniCPM-Embedding-Light)
+- 重排模型：[MiniCPM-Reranker-Light](https://huggingface.co/openbmb/MiniCPM-Reranker-Light)
 - 领域自适应RAG框架：[UltraRAG](https://github.com/openbmb/UltraRAG)
 
-**UltraRAG-Reranker** is a bilingual & cross-lingual text re-ranking model developed by ModelBest Inc. , THUNLP and NEUIR , featuring:
+**MiniCPM-Reranker-Light** is a bilingual & cross-lingual text re-ranking model developed by ModelBest Inc. , THUNLP and NEUIR , featuring:
 
 - Exceptional Chinese and English re-ranking capabilities.
 - Outstanding cross-lingual re-ranking capabilities between Chinese and English.
 - Long-text support (up to 8192 tokens).
 
-UltraRAG-Reranker is trained based on [MiniCPM-1B-sft-bf16](https://huggingface.co/openbmb/MiniCPM-1B-sft-bf16) and incorporates bidirectional attention in its architecture. The model underwent multi-stage training using approximately 6 million training examples, including open-source, synthetic, and proprietary data.
+MiniCPM-Reranker-Light is trained based on [MiniCPM-1B-sft-bf16](https://huggingface.co/openbmb/MiniCPM-1B-sft-bf16) and incorporates bidirectional attention in its architecture. The model underwent multi-stage training using approximately 6 million training examples, including open-source, synthetic, and proprietary data.
 
 We also invite you to explore the UltraRAG series:
 
-- Retrieval Model: [UltraRAG-Embedding](https://huggingface.co/openbmb/UltraRAG-Embedding)
-- Re-ranking Model: [UltraRAG-Reranker](https://huggingface.co/openbmb/UltraRAG-Reranker)
+- Retrieval Model: [MiniCPM-Embedding-Light](https://huggingface.co/openbmb/MiniCPM-Embedding-Light)
+- Re-ranking Model: [MiniCPM-Reranker-Light](https://huggingface.co/openbmb/MiniCPM-Reranker-Light)
 - Domain Adaptive RAG Framework: [UltraRAG](https://github.com/openbmb/UltraRAG)
 
 
@@ -52,7 +52,7 @@ We also invite you to explore the UltraRAG series:
 
 本模型支持指令，输入格式如下：
 
-UltraRAG-Reranker supports instructions in the following format:
+MiniCPM-Reranker-Light supports instructions in the following format:
 
 ```
 <s>Instruction: {{ instruction }} Query: {{ query }}</s>{{ document }}
@@ -72,7 +72,7 @@ For example:
 
 也可以不提供指令，即采取如下格式：
 
-UltraRAG-Reranker also works in instruction-free mode in the following format:
+MiniCPM-Reranker-Light also works in instruction-free mode in the following format:
 
 ```
 <s>Query: {{ query }}</s>{{ document }}
@@ -96,7 +96,7 @@ transformers==4.37.2
 from transformers import AutoModelForSequenceClassification
 import torch
 
-model_name = "OpenBMB/UltraRAG-Reranker"
+model_name = "OpenBMB/MiniCPM-Reranker-Light"
 
 # model = AutoModelForSequenceClassification.from_pretrained(model_name, trust_remote_code=True,attn_implementation="flash_attention_2", torch_dtype=torch.float16).to("cuda")
 model.eval()
@@ -120,7 +120,7 @@ from sentence_transformers import CrossEncoder
 from transformers import LlamaTokenizer
 import torch
 
-model_name = "OpenBMB/UltraRAG-Reranker"
+model_name = "OpenBMB/MiniCPM-Reranker-Light"
 model = CrossEncoder(model_name,max_length=1024,trust_remote_code=True, automodel_args={"torch_dtype": torch.float16})
 # You can also use the following code to use flash_attention_2
 #model = CrossEncoder(model_name,max_length=1024,trust_remote_code=True, automodel_args={"attn_implementation":"flash_attention_2","torch_dtype": torch.float16})
@@ -157,7 +157,7 @@ INSTRUCTION = "Query:"
 query = f"{INSTRUCTION} {query}"
 
 array = AsyncEngineArray.from_args(
-  [EngineArgs(model_name_or_path = "OpenBMB/UltraRAG-Reranker", engine="torch", dtype="float16", bettertransformer=False, trust_remote_code=True, model_warmup=False)]
+  [EngineArgs(model_name_or_path = "OpenBMB/MiniCPM-Reranker-Light", engine="torch", dtype="float16", bettertransformer=False, trust_remote_code=True, model_warmup=False)]
 )
 
 async def rerank(engine: AsyncEmbeddingEngine): 
@@ -172,7 +172,7 @@ asyncio.run(rerank(array[0])) # [(RerankReturnType(relevance_score=0.017917344,
 
 ```python
 from FlagEmbedding import FlagReranker
-model_name = "OpenBMB/UltraRAG-Reranker"
+model_name = "OpenBMB/MiniCPM-Reranker-Light"
 model = FlagReranker(model_name, use_fp16=True, query_instruction_for_rerank="Query: ", trust_remote_code=True)
 # You can hack the __init__() method of the FlagEmbedding BaseReranker class to use flash_attention_2 for faster inference
 #  self.model = AutoModelForSequenceClassification.from_pretrained(
@@ -211,7 +211,7 @@ We re-rank top-100 docments from `bge-large-zh-v1.5` in C-MTEB/Retrieval and fro
 | bge-reranker-v2-gemma      | 71.74             | 60.71         |
 | bge-reranker-v2.5-gemma2   | -                 | 63.67    |
 | MiniCPM-Reranker                 | 76.79         | 61.32        |
-| UltraRAG-Reranker                | 76.19      |	61.34       |
+| MiniCPM-Reranker-Light                | 76.19      |	61.34       |
 
 ### 中英跨语言重排序结果 CN-EN Cross-lingual Re-ranking Results
 
@@ -226,14 +226,14 @@ We re-rank top-100 documents from `bge-m3` (Dense).
 | bge-reranker-v2-m3                 | 69.75              | 40.98              | 49.67              |
 | gte-multilingual-reranker-base     | 68.51              | 38.74              | 45.3              |
 | MiniCPM-Reranker                         | 71.73          | 43.65          | 50.59          |
-| UltraRAG-Reranker                        | 71.34       | 46.04       | 51.86       |
+| MiniCPM-Reranker-Light                        | 71.34       | 46.04       | 51.86       |
 
 ## 许可证 License
 
 - 本仓库中代码依照 [Apache-2.0 协议](https://github.com/OpenBMB/MiniCPM/blob/main/LICENSE)开源。
-- UltraRAG-Reranker 模型权重的使用则需要遵循 [MiniCPM 模型协议](https://github.com/OpenBMB/MiniCPM/blob/main/MiniCPM%20Model%20License.md)。
-- UltraRAG-Reranker 模型权重对学术研究完全开放。如需将模型用于商业用途，请填写[此问卷](https://modelbest.feishu.cn/share/base/form/shrcnpV5ZT9EJ6xYjh3Kx0J6v8g)。
+- MiniCPM-Reranker-Light 模型权重的使用则需要遵循 [MiniCPM 模型协议](https://github.com/OpenBMB/MiniCPM/blob/main/MiniCPM%20Model%20License.md)。
+- MiniCPM-Reranker-Light 模型权重对学术研究完全开放。如需将模型用于商业用途，请填写[此问卷](https://modelbest.feishu.cn/share/base/form/shrcnpV5ZT9EJ6xYjh3Kx0J6v8g)。
 
 * The code in this repo is released under the [Apache-2.0](https://github.com/OpenBMB/MiniCPM/blob/main/LICENSE) License. 
-* The usage of UltraRAG-Reranker model weights must strictly follow [MiniCPM Model License.md](https://github.com/OpenBMB/MiniCPM/blob/main/MiniCPM%20Model%20License.md).
-* The models and weights of UltraRAG-Reranker are completely free for academic research. After filling out a ["questionnaire"](https://modelbest.feishu.cn/share/base/form/shrcnpV5ZT9EJ6xYjh3Kx0J6v8g) for registration, UltraRAG-Reranker weights are also available for free commercial use.
+* The usage of MiniCPM-Reranker-Light model weights must strictly follow [MiniCPM Model License.md](https://github.com/OpenBMB/MiniCPM/blob/main/MiniCPM%20Model%20License.md).
+* The models and weights of MiniCPM-Reranker-Light are completely free for academic research. After filling out a ["questionnaire"](https://modelbest.feishu.cn/share/base/form/shrcnpV5ZT9EJ6xYjh3Kx0J6v8g) for registration, MiniCPM-Reranker-Light weights are also available for free commercial use.