Skip to content

Latest commit

 

History

History
21 lines (13 loc) · 890 Bytes

dialog.md

File metadata and controls

21 lines (13 loc) · 890 Bytes

Dialog

Dialogue is notoriously hard to evaluate. Past approaches have used human evaluation.

Dialog state tracking

Dialogue state tacking consists of determining at each turn of a dialog the full representation of what the user wants at that point in the dialog, which contains a goal constraint, a set of requested slots, and the user's dialog act.

Second dialog state tracking challenge

For goal-oriented dialogue, the dataset of the second dialog state tracking challenge (DSTC2) is a common evaluation dataset. The DSTC2 focuses on the restaurant search domain. Models are evaluated based on accuracy on both individual and joint slot tracking.

{% include table.html results=site.data.dialog scores='Area,Food,Price,Joint' %}

{% include chart.html results=site.data.dialog score='Joint' %}

Go back to the README