readme format

Author: Achazwl

README-ZH.md

@@ -33,18 +33,18 @@
 
 ## 最新动态
 
-- 2022/03/16 [0.1.0](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter 公开发布了第一个稳定版本, 修复了一些模型性能上和显存占用上的问题.
-- 2022/03/21 [0.0.1-beta](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter 公开发布了第一个 beta 版本.
+- 2022/03/31 [**ModelCenter 0.1.0**](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter 公开发布了第一个稳定版本, 修复了一些模型性能上和显存占用上的问题.
+- 2022/03/21 [**ModelCenter 0.0.1-beta**](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter 公开发布了第一个 beta 版本.
 
 ## 总览
 
 ModelCenter 基于 [OpenBMB/BMTrain](https://github.com/OpenBMB/BMTrain/) 实现了一系列经典的预训练语言模型。 ModelCenter 在模型实现上的宗旨是 高效、低资源与高可用性, 并且能够支持分布式的训练.
 
 我们的主要优势有：
 
-- 易用性：相比 Deepspeed, Megatron, 我们拥有更好更灵活的封装，且配置 python 环境容易, 训练代码与 PyTorch 风格统一。
-- 更高效的显存利用：模型占用显存较大时，可能会导致 GPU 的计算能力未被充分使用时显存占用就已经跑满。我们的实现可以将显存占用降低数倍，进而使用更大的 batch-size 对 GPU 的计算能力进行更充分的利用。
-- 低资源的高效分布式训练：在 [OpenBMB/BMTrain](https://github.com/OpenBMB/BMTrain/) 的支持下，我们能够将 ZeRO3 的优化轻易地扩展至各大预训练语言模型，并在分布式训练的通信和调度上作出优化。
+- **易用性**。相比 Deepspeed, Megatron, 我们拥有更好更灵活的封装，且配置 python 环境容易, 训练代码与 PyTorch 风格统一。
+- **更高效的显存利用**。模型占用显存较大时，可能会导致 GPU 的计算能力未被充分使用时显存占用就已经跑满。我们的实现可以将显存占用降低数倍，进而使用更大的 batch-size 对 GPU 的计算能力进行更充分的利用。
+- **低资源的高效分布式训练**。在 [OpenBMB/BMTrain](https://github.com/OpenBMB/BMTrain/) 的支持下，我们能够将 ZeRO3 的优化轻易地扩展至各大预训练语言模型，并在分布式训练的通信和调度上作出优化。
 
 ## 文档
 

README.md

@@ -33,18 +33,18 @@
 
 ## What's New
 
-- 2022/03/16 [0.1.0](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter has publicly released the first stable version, which fixes some bugs in model performance and GPU memory usage.
-- 2022/03/21 [0.0.1-beta](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter has publicly released the first beta version.
+- 2022/03/32 [**ModelCenter 0.1.0**](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter has publicly released the first stable version, which fixes some bugs in model performance and GPU memory usage.
+- 2022/03/21 [**ModelCenter 0.0.1-beta**](https://github.com/OpenBMB/ModelCenter/releases/tag/v0.0.1-beta) ModelCenter has publicly released the first beta version.
 
 ## Overview
 
 ModelCenter implements pre-trained language models (PLMs) based on [OpenBMB/BMTrain](https://github.com/OpenBMB/BMTrain/) backend. ModelCenter supports Efficient, Low-Resource, Extendable model usage and distributed training.
 
 Our main advantages are:
 
-- Easy to use: Compared to Deepspeed and Megatron, we have better and more flexible code-packaging and easy to configure python environments, and the training code is uniform with PyTorch style.
-- More efficient memory utilization: Models with large memory footprints can cause OOM (out of memory) before the computational power of the GPU is fully utilized. Our implementation reduces the memory footprint by several times, allowing more efficient use of the GPU's computational power with a larger batch size.
-- Efficient distributed training with low resources: With the support of [OpenBMB/BMTrain](https://github.com/OpenBMB/BMTrain/), we are able to easily extend ZeRO3's optimization to any PLMs, and we optimize communication and time scheduling for faster distributed training.
+- **Easy to use**. Compared to Deepspeed and Megatron, we have better and more flexible code-packaging and easy to configure python environments, and the training code is uniform with PyTorch style.
+- **More efficient memory utilization**. Models with large memory footprints can cause OOM (out of memory) before the computational power of the GPU is fully utilized. Our implementation reduces the memory footprint by several times, allowing more efficient use of the GPU's computational power with a larger batch size.
+- **Efficient distributed training with low resources**. With the support of [OpenBMB/BMTrain](https://github.com/OpenBMB/BMTrain/), we are able to easily extend ZeRO3's optimization to any PLMs, and we optimize communication and time scheduling for faster distributed training.
 
 ## Documentation