[Adapt] [Seminar]Experiments in NLP(II)

贾琪 jia_qi_0217 at 163.com
Tue Nov 12 19:48:02 CST 2019


Hi Adapters,


Two weeks ago, we talked about the tips for reproducing SOTA baselines and experimental designs. This week, I will follow the previous topic and give some suggestions on analyzing of experiment results to make the results reproducible and convincible. Besides, due to the popularity of natural language generation tasks and the demand of human evaluation for such tasks, I will talk about some key points and difficulties when implementing the human labeling task. 


TIme: 16:30 13/11/2019 
Venue: SEIEE 3-414


Best,
Angel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20191112/487ca56f/attachment.html>


More information about the Adapt mailing list