Curriculum Learning for Natural Answer Generation – 第2页 – NLPIR自然语言处理与信息检索共享平台

自然语言处理与信息检索共享平台 自然语言处理与信息检索共享平台

Curriculum Learning for Natural Answer Generation


Last week, BaohuaZhang gave a presentation about the paper, Curriculum Learning for Natural Answer Generation, and shared some opinion on it.


After the presentation, several questions were asked. The Q&As are listed as following:

Q1: Does the Chinese and English have the same model? A1: No, the model is only for Chinese.

Q2: Is the accuracy of this model not higher the basic model?
A2: Most of the model’s anwsers are right, and the answers are more natural and closer to our human’s expressions.

Q3: This paper is about curriculum leaning, what’s the difference between it and general deep learning?
A3: Curriculum learning starts from easy instances and then gradually handles harder ones. In this paper the author uses instance scheduler, the instance contains target instances and common instance, and trains the model from common instances to target instances.

Q4: Which model does NAG use?
A4: NAG uses gold topic entities to simplify the model.

You May Also Like

About the Author: nlpvv