# 期末彙整
## Papers
* [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/abs/1810.04805)
* 往後的NLP任務都使用Transfer Learning
* [Towards a Human-like Open-Domain Chatbot](https://arxiv.org/abs/2001.09977)
* Google 的開放域聊天機器人"Meena"
* [PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable](https://arxiv.org/abs/1910.07931)
* [PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning](https://arxiv.org/abs/2006.16779)
* 百度的開放域聊天機器人"PLATO"

---

* [基於對話情境及同理分析於條件轉換器之同理回應生成](https://etdquery.lib.ncku.edu.tw/etds_query/etd_detail.php?UWYPcgBhDnQGaVE8XQIKNFFjCDgANVYtA2ICMgM3CT9SMg00UGdbZAAzA2AMNwA8Dz4GN1NgCmw%3D)
* 架構中分別有情緒偵測、對話主題偵測、情境偵測、回應生成
* 情緒偵測和對話主題偵測用BERT預訓練，情境偵測用SBERT預訓練，再用Conditional Transformer做回應生成

---

* Knowledge Graph(知識圖譜)
研究是否可用在ChatBot作為個性化知識圖譜