Papers by Shih-Hung Wu
Learning the Human Judgment for the Automatic Evaluation of Chatbot (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing evaluation methods for dialogue systems rely on human judges to label quality of generated text. |
| Approach: | They propose a machine learning approach to reduce the effort of human evaluation by learning the human judgment on comparing two generative dialogue systems. |
| Outcome: | The proposed method reduces the effort of human evaluation by learning which generative models is better in each dialog context. |