CoCo: Controllable Counterfactuals for Evaluating Dialogue State Trackers
ICLR 2021, 2021.
CoCo is a principled method to more flexibly evaluate the robustness of the DST component of TOD systems.
Dialogue state trackers have made significant progress on benchmark datasets, but their generalization capability to novel and realistic scenarios beyond the held-out conversations is less understood. We propose controllable counterfactuals (CoCo) to bridge this gap and evaluate dialogue state tracking (DST) models on novel scenarios, i...More
PPT (Upload PPT)