Task-Oriented Conversation Generation Using Heterogeneous Memory Networks
EMNLP 2019 錄用長文 Abstract 如何將外部知識庫與對話模型結合起來是一個重要的問題。傳統上人們是使用了Memory Network。然而當面對多種來源的、異構的info時，Mem對這些info的處理並不好（我理解的是權重的問題）。因此提出了HMN，它能同步的使用user utterances,dialogue history和backgroud knowledge tuple
相關文章
- 1. 《Emotional Chatting Machine: Emotional Conversation Generation with Internal and External Memory》筆記
- 2. Emotional Chatting Machine: Emotional Conversation Generation with Internal and External Memory
- 3. Summary - Video Object Segmentation using Space-Time Memory Networks
- 4. Emotion Recognition Using Graph Convolutional Networks
- 5. 【ICJAI2018】Commonsense Knowledge Aware Conversation Generation with Graph Attention
- 6. 關於conversation generation的論文筆記
- 7. 【EMNLP2019】Knowledge Aware Conversation Generation with Explainable Reasoning over Augmented Graphs
- 8. Memory Networks
- 9. Review networks for caption generation
- 10. metapath2vec: Scalable Representation Learning for Heterogeneous Networks
- 更多相關文章...
- • Maven Web 應用 - Maven教程
- • Maven NetBeans - Maven教程
- • RxJava操作符（六）Utility
- • C# 中 foreach 遍歷的用法
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. 《Emotional Chatting Machine: Emotional Conversation Generation with Internal and External Memory》筆記
- 2. Emotional Chatting Machine: Emotional Conversation Generation with Internal and External Memory
- 3. Summary - Video Object Segmentation using Space-Time Memory Networks
- 4. Emotion Recognition Using Graph Convolutional Networks
- 5. 【ICJAI2018】Commonsense Knowledge Aware Conversation Generation with Graph Attention
- 6. 關於conversation generation的論文筆記
- 7. 【EMNLP2019】Knowledge Aware Conversation Generation with Explainable Reasoning over Augmented Graphs
- 8. Memory Networks
- 9. Review networks for caption generation
- 10. metapath2vec: Scalable Representation Learning for Heterogeneous Networks