Our paper titled Credibility-Aware Multi-Modal Fusion Using Probabilistic Circuits has been accepted to AISTATS 2025

Our paper titled “Credibility-Aware Multi-Modal Fusion Using Probabilistic Circuits” has been accepted to the 28th International Conference on Artificial Intelligence and Statistics (AISTATS) 2025! This paper explores the notion of credibility in noisy, multi-source domains that require understanding the reliability of each data source, and proposes a combination function that uses probabilistic circuits (PCs) to combine predictive distributions over individual modalities. This work also defines a probabilistic measure to evaluate the credibility of each modality via inference queries over the PC. To learn more about this work, please click here