Using Domain Knowledge to Guide Dialog Structure Induction via Neural Probabilistic Soft Logic

arXiv:2403.17853v1 Announce Type: cross
Abstract: Dialog Structure Induction (DSI) is the task of inferring the latent dialog structure (i.e., a set of dialog states and their temporal transitions) of a given goal-oriented dialog. It is a critical component for modern dialog system design and discourse analysis. Existing DSI approaches are often purely data-driven, deploy models that infer latent states without access to domain knowledge, underperform when the training corpus is limited/noisy, or have difficulty when test dialogs exhibit distributional shifts …

abstract analysis arxiv cs.cl cs.lg design dialog discourse domain domain knowledge guide knowledge logic modern set temporal transitions type via