diff --git a/README.md b/README.md
index ef5281f..868b030 100644
--- a/README.md
+++ b/README.md
@@ -33,4 +33,47 @@
 | TBD  | (直播-Discussion)  <br> 代码实战：基于LSTM-CRF的命名实体识别||||||
 | TBD  | (直播-Discussion)  TBD||||||
 | TBD  | (直播-Discussion)  TBD||||||
-|PART 2 深度学习与预训练模型|
\ No newline at end of file
+|PART 2 深度学习与预训练模型|
+|TBD  | (直播-Lecture10)   <br> CRF模型（2），深度学习基础|||[Log-linear models </br>and conditional </br>random fields</br>](http://cseweb.ucsd.edu/~elkan/250B/CRFs.pdf)，</br>[An Introduction </br>to Conditional</br> Random Fields</br>](http://homepages.inf.ed.ac.uk/csutton/publications/crftut-fnt.pdf)，</br>[Log-Linear Models and Conditional</br> Random Field</br>](http://cseweb.ucsd.edu/~elkan/250Bfall2007/loglinear.pdf)，</br>[Generative Learning algorithms</br>](http://cs229.stanford.edu/notes/cs229-notes2.pdf),[网址](http://videolectures.net/cikm08_elkan_llmacrf/)|||
+|TBD  | (直播-Paper)       <br> |||||
+| TBD  | (直播-Discussion)  <br> GPU的使用与环境搭建 +  基于pytorch的简单的神经网络搭建|||[阅读资料](http://47.94.6.102/NLPCamp6/course-info/blob/master/%E8%AF%BE%E4%BB%B6/0517GPU+pytorch%E9%98%85%E8%AF%BB%E8%B5%84%E6%96%99cuda_c_chinese.pdf)|||
+| TBD  | (直播-Discussion)  <br> 对话系统技术概览||||||
+|TBD  | (直播-Lecture11)   <br> RNN, LSTM，梯度问题 |||||
+| TBD | (直播-Paper)       <br> |||||
+| TBD  | (直播-Discussion)  <br> Pytorch讲解||||||
+| TBD  | (直播-Lecture12)   <br> Seq2Seq, Attention, Pointer Network ||||||
+| TBD | (直播-Paper)       <br> ||||||
+| TBD   | (直播-Discussion)  <br> Introduction to Transfer Learing||||||
+| TBD  | (直播-Discussion)  <br> LSTM的实现（源码讲解） ||||||
+| TBD  | (直播-Lecture13)   <br> Transformer, BERT ||||||
+|TBD  | (直播-Paper)       <br> ||||||
+| TBD  | (直播-Discussion)  <br> 基于Transformer的机器翻译||||||
+| TBD  | (直播-Discussion)  <br> BERT的训练与实战||||||
+| TBD | (直播-Lecture14)   <br> GPT, XLNet |||||
+| TBD  | (直播-Discussion)  <br> XLNET应用在文本分类和QA系统||||||
+| TBD | (直播-Discussion)  <br> XLNET源码讲解||||||
+| TBD  | (直播-Paper)       <br> ||||||
+| PART 3  信息抽取与图挖掘相关|
+| TBD  | (直播-Lecture15)   <br> 信息抽取(1) |利用传统方法论解决命名实体识别，实体统一，实体消歧，关系抽取问题|||||
+| TBD | (直播-Paper)       <br> ||||||
+| TBD  | (直播-Discussion)  <br> 命名实体识别代码实战：BERT-BILSTM-CRF||||||
+| TBD   | (直播-Discussion)  <br> ALBERT |||[ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942)||
+| TBD   | (直播-Discussion)  <br> 项目二讲解||||||
+| TBD  | (直播-Lecture16)   <br> 信息抽取（2）|利用深度学习解决命名实体识别，实体统一，实体消歧，关系抽取问题|)||||
+| TBD  | (直播-Paper)       <br> ||||
+| TBD  | (直播-Discussion)  <br> 依存文法分析（Dependency Parsing) |||||
+| TBD   | (直播-Discussion)  <br> 句法分析（Parsing)和CKY算法 |||||
+| TBD  | (直播-Lecture17)   <br> 知识图谱 |知识图的概念，搭建，应用场景||||
+| TBD  | (直播-Paper)       <br> ||||||
+| TBD   | (直播-Discussion)  <br> 知识图谱在推荐系统中的应用 |||||
+| TBD  | (直播-Discussion)  <br> project3 项目讲解 |||||
+| TBD  | (直播-Lecture18)   <br> 图神经网络 |图卷积审计网络，GraphSage, GAT||||
+| PART 4  概率图模型|
+| TBD  | (直播-Lecture20)   <br> 概率图模型-1 |贝叶斯推理，LDA模型||||
+| TBD | (直播-Discussion)   <br> MCMC之Metroplis Hasting算法 |||||
+| TBD | (直播-Discussion)   <br> Bayesian Neural Network |||||
+| TBD | (直播-Paper)   <br>  |||[Dropout as a Bayesian Approximation:Representing Model Uncertainty in Deep Learning](https://arxiv.org/pdf/1506.02142.pdf)||
+| TBD | (直播-Lecture21)   <br> 概率图模型-2 |吉布斯采样、变分法||||
+
+
+