update readme zh

README.md

@@ -1073,7 +1073,7 @@ effectively harnessing textual data and labels from diverse downstream
 tasks. In addition, Piccolo2 scales up the embedding dimension and uses
 MRL training to support more flexible vector dimensions.
 
-## Model Hightlights
+## 💡 Model Hightlights
 The main feature of piccolo2 is that it uses a multi-task hybrid loss during training.   
 For retrieval/sorting tasks, we use the standard InfoNCE with in-batch-negative:
 <p align='left'>
@@ -1092,7 +1092,7 @@ it can easily lead to conflict training targets:
 <img src='assets/3.png' width='400' height='80'>
 </p>
 
-## Experiments and Results
+## 📃 Experiments and Results
 Piccolo2 primarily focuses on the downstream general finetune paradigm. Our open source model uses [stella-v3.5](https://huggingface.co/infgrad/stella-mrl-large-zh-v3.5-1792d) as initialization and trained about 2500 steps on 32 GPUS. For more implementation details, please refer to our [technical report](https://arxiv.org/abs/2405.06932).
 
 | Model Name | Model Size (GB) | Dimension | Sequence Length |  Classification (9) | Clustering (4) | Pair Classification (2) | Reranking (4) | Retrieval (8) | STS (8) | Average (35) |
@@ -1102,7 +1102,7 @@ Piccolo2 primarily focuses on the downstream general finetune paradigm. Our open
 | [acge-text-embedding](https://huggingface.co/aspire/acge_text_embedding) |1.21 | 1792 | 512 | 72.75 | 58.7 | 87.84 | 67.98 | 72.93 | 62.09 |  69.07 |
 
 
-## Usage
+## 🔨 Usage
 The piccolo model can be easily accessed in the sentence-transformer package:
 ```python
 # for s2s/s2p dataset, you can use piccolo as below

README_zh.md

@@ -0,0 +1,79 @@
+[EN](README.md) | [简体中文](README_zh.md) 
+
+**新闻**
+
+**[2024-05-14]**   
+我们目前已经发布了模型权重、训练代码和技术报告，欢迎大家关注。   
+我们的训练代码放在了github上: https://github.com/hjq133/piccolo-embedding   
+对于训练细节，可以参考我们的技术报告: https://arxiv.org/abs/2405.06932
+
+
+**[2024-04-22]** 
+
+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。  
+
+## Piccolo-large-zh-v2
+
+piccolo-large-zh-v2是商汤研究院通用模型组开发的中文嵌入模型。 Piccolo此次的升级版本旨在关注通用的下游微调方法。 Piccolo2主要通过利用高效的多任务混合损失训练方法，有效地利用来自不同下游的文本数据和标签。 此外，Piccolo2扩大了嵌入维度，同时使用MRL训练来支持更灵活的向量维度。
+
+## 💡 模型亮点
+piccolo2的主要特点是在训练过程中使用了多任务混合损失。  
+对于检retrieval/reranking任务，我们使用带有批内负样本的InfoNCE:
+<p align='left'>
+<img src='assets/1.png' width='400' height='80'>
+</p>
+
+对于 sts/pair classification任务，我们使用排序损失:cosent loss。在具有细粒度标签的数据集上（比如有相似度的score），排序损失通常被证明表现更好:
+<p align='left'>
+<img src='assets/2.png' width='450' height='90'>
+</p>
+
+对于分类/聚类任务，我们通过将文本和其语义标签视为正负对，将数据集转换为三元组的格式来进行训练。我们同样采用InfoNCE对其进行优化。但这类任务不能再使用批内负样本，因为很容易导致训练目标的冲突：
+<p align='left'>
+<img src='assets/3.png' width='400' height='80'>
+</p>
+
+## 📃 实验和结果
+Piccolo2主要关注在一种通用的下游微调范式。我们的开源模型使用了[stella-v3.5](https://huggingface.co/infgrad/stella-mrl-large-zh-v3.5-1792d)作为初始化，在32张A100上训练了2500 step，对于更多的实现细节，可以参考我们的 [技术报告](https://arxiv.org/abs/2405.06932), 以及[训练代码](https://github.com/hjq133/piccolo-embedding)
+
+| Model Name | Model Size (GB) | Dimension | Sequence Length |  Classification (9) | Clustering (4) | Pair Classification (2) | Reranking (4) | Retrieval (8) | STS (8) | Average (35) |
+|:----:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
+| [**piccolo-large-zh-v2**](https://huggingface.co/sensenova/piccolo-large-zh-v2) | 1.21 | 1792 | 512 | 74.59 | 62.17 | 90.24 | 70 | 74.36 | 63.5 | 70.95 |
+| [gte-Qwen1.5-7B-instruct](https://huggingface.co/Alibaba-NLP/gte-Qwen1.5-7B-instruct)| 26.45 | 32768 |4096 | 73.35 |  67.08 |  88.52 |  66.38 | 70.62 | 62.32 |  69.56|
+| [acge-text-embedding](https://huggingface.co/aspire/acge_text_embedding) |1.21 | 1792 | 512 | 72.75 | 58.7 | 87.84 | 67.98 | 72.93 | 62.09 |  69.07 |
+
+
+## 🔨 使用方法
+在sentence-transformer中使用piccolo:
+```python
+# for s2s/s2p dataset, you can use piccolo as below
+from sklearn.preprocessing import normalize
+from sentence_transformers import SentenceTransformer
+sentences = ["数据1", "数据2"]
+matryoshka_dim=1792 # support 256, 512, 768, 1024, 1280, 1536, 1792
+model = SentenceTransformer('sensenova/piccolo-large-zh-v2')
+embeddings_1 = model.encode(sentences, normalize_embeddings=False)
+embeddings_2 = model.encode(sentences, normalize_embeddings=False)
+embeddings_1 = normalize(embeddings_1[..., :matryoshka_dim], norm="l2", axis=1)
+embeddings_2 = normalize(embeddings_2[..., :matryoshka_dim], norm="l2", axis=1)
+similarity = embeddings_1 @ embeddings_2.T
+```
+
+## 🤗 **Model List**
+| Model|Language|Description|prompt|
+|:-|:-:|:-:|:--:|
+| [sensenova/piccolo-large-zh-v2](https://huggingface.co/sensenova/piccolo-large-zh-v2)                   |    Chinese    | version2: finetuning with multi-task hybrid loss training | None |
+| [sensenova/piccolo-large-zh](https://huggingface.co/sensenova/piccolo-large-zh)                   |    Chinese    | version1: pretrain under 400 million chinese text pair | '查询'/'结果' |
+| [sensenova/piccolo-base-zh](https://huggingface.co/sensenova/piccolo-base-zh)                   |    Chinese    | version1: pretrain under 400 million chinese text pair | '查询'/'结果' |
+
+
+## Citation
+如果我们的技术报告、模型或训练代码对您有帮助，请像下面这样引用我们的论文，或者在 github、 Huggingface 上给一个 Star！
+```bibtex
+@misc{2405.06932,
+Author = {Junqin Huang and Zhongjie Hu and Zihao Jing and Mengya Gao and Yichao Wu},
+Title = {Piccolo2: General Text Embedding with Multi-task Hybrid Loss Training},
+Year = {2024},
+Eprint = {arXiv:2405.06932},
+}
+```