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.
|Model||Area||Food||Price||Joint||Paper / Source||Code|
|Liu et al. (2018)||90||84||92||72||Dialogue Learning with Human Teaching and Feedback in End-to-End Trainable Task-Oriented Dialogue Systems|
|Neural belief tracker by Mrkšić et al. (2017)||90||84||94||72||Neural Belief Tracker: Data-Driven Dialogue State Tracking|
|RNN by Henderson et al. (2014)||92||86||86||69||Robust dialog state tracking using delexicalised recurrent neural networks and unsupervised gate|
Liu et al. (2018)
Mrkšić et al. (2017)
Henderson et al. (2014)