What Did You Say? Task-Oriented Dialog Datasets Are Not Conversational!?
This highlights a critical flaw in widely used datasets for virtual assistant development, potentially hindering progress in conversational AI.
The paper identified that major task-oriented dialog datasets (MultiWOZ, SGD, SMCalFlow) are largely non-conversational, as dialog state tracking can often be done using only the current user utterance, ignoring history, with less than 4% of MultiWOZ's turns and 10% of SGD's turns being conversational.
High-quality datasets for task-oriented dialog are crucial for the development of virtual assistants. Yet three of the most relevant large scale dialog datasets suffer from one common flaw: the dialog state update can be tracked, to a great extent, by a model that only considers the current user utterance, ignoring the dialog history. In this work, we outline a taxonomy of conversational and contextual effects, which we use to examine MultiWOZ, SGD and SMCalFlow, among the most recent and widely used task-oriented dialog datasets. We analyze the datasets in a model-independent fashion and corroborate these findings experimentally using a strong text-to-text baseline (T5). We find that less than 4% of MultiWOZ's turns and 10% of SGD's turns are conversational, while SMCalFlow is not conversational at all in its current release: its dialog state tracking task can be reduced to single exchange semantic parsing. We conclude by outlining desiderata for truly conversational dialog datasets.