The Open-World Lottery Ticket Hypothesis for OOD Intent Classification

arXiv:2210.07071v2 Announce Type: replace
Abstract: Most existing methods of Out-of-Domain (OOD) intent classification rely on extensive auxiliary OOD corpora or specific training paradigms. However, they are underdeveloped in the underlying principle that the models should have differentiated confidence in In- and Out-of-domain intent. In this work, we shed light on the fundamental cause of model overconfidence on OOD and demonstrate that calibrated subnetworks can be uncovered by pruning the overparameterized model. Calibrated confidence provided by the subnetwork can better distinguish …

abstract arxiv classification confidence cs.cl domain however hypothesis light lottery lottery ticket hypothesis open-world training type work world