[NLG]A Working Memory Model for Task-oriented Dialog Response Generation
总述： memNN常用于在对话中引入KB的知识，然而之前的工作大多数把dialogue history和KB的memory混在一起，作者借鉴了心理学的研究。有个work memory（类似于电脑内存的感觉）可以和两个记忆做交互，一个知识记忆（semantic memory），一个历史记忆（episodic memory）.然后作者就设计了一个模型，将对话历史作为历史记忆，将KB结构化知识作为知识记
相关文章
- 1. Tips for Generation in Encoder-Decoder model
- 2. Guiding the Long-Short Term Memory Model for Image Caption Generation
- 3. 知识驱动对话-Learning to Select Knowledge for Response Generation in Dialog Systems-阅读笔记
- 4. [Coling2018]SGM：Sequence Generation Model for Multi-Label Classification
- 5. Task-Oriented Conversation Generation Using Heterogeneous Memory Networks
- 6. linux 3个内存模型(flat memory model、discontiguous memory model、sparse memory model)
- 7. 带你走进 JVM Memory Model
- 8. 论文笔记：基于外部知识的会话模型Learning to Select Knowledge for Response Generation in Dialog Systems
- 9. CUDA ---- Memory Model
- 10. Neural Response Generation via GAN with an Approximate Embedding Layer
- 更多相关文章...
- • ASP Response 对象 - ASP 教程
- • Swift for 循环 - Swift 教程
- • Flink 数据传输及反压详解
- • Docker容器实战(六) - 容器的隔离与限制
相关标签/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。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. Tips for Generation in Encoder-Decoder model
- 2. Guiding the Long-Short Term Memory Model for Image Caption Generation
- 3. 知识驱动对话-Learning to Select Knowledge for Response Generation in Dialog Systems-阅读笔记
- 4. [Coling2018]SGM：Sequence Generation Model for Multi-Label Classification
- 5. Task-Oriented Conversation Generation Using Heterogeneous Memory Networks
- 6. linux 3个内存模型(flat memory model、discontiguous memory model、sparse memory model)
- 7. 带你走进 JVM Memory Model
- 8. 论文笔记：基于外部知识的会话模型Learning to Select Knowledge for Response Generation in Dialog Systems
- 9. CUDA ---- Memory Model
- 10. Neural Response Generation via GAN with an Approximate Embedding Layer