MultiWOZ 2.0

AbstractRecent studies in dialogue state tracking (DST) leverage historical information to determine states which are generally represented as slot-value pairs. However, most of them have limitations to efficiently exploit relevant …...

AbstractRecent works in dialogue state tracking (DST) focus on an open vocabulary-based setting to resolve scalability and generalization issues of the predefined ontology-based approaches. However, they are inefficient in that …...

AbstractThe goal of dialogue state tracking (DST) is to predict the current dialogue state given all previous dialogue contexts. Existing approaches generally predict the dialogue state at every turn from …...

AbstractThe recent success of reinforcement learning (RL) in solving complex tasks is often attributed to its capacity to explore and exploit an environment. Sample efficiency is usually not an issue …...

AbstractThe MultiWOZ 2.0 dataset has greatly boosted the research on dialogue state tracking (DST). However, substantial noise has been discovered in its state annotations. Such noise brings about huge challenges …...