Credibility-Aware Multi-Modal Fusion Using Probabilistic Circuits

arXiv:2403.03281v1 Announce Type: new
Abstract: We consider the problem of late multi-modal fusion for discriminative learning. Motivated by noisy, multi-source domains that require understanding the reliability of each data source, we explore the notion of credibility in the context of multi-modal fusion. We propose a combination function that uses probabilistic circuits (PCs) to combine predictive distributions over individual modalities. We also define a probabilistic measure to evaluate the credibility of each modality via inference queries over the PC. Our experimental …

abstract arxiv combination context cs.ai cs.lg data domains explore function fusion modal multi-modal notion pcs reliability type understanding