GDP predicts by sampling candidate outputs from a conditional generative model and selecting the one with the lowest average loss; its excess risk is bounded by generation error plus a vanishing sampling term.
Baltrušaitis, C
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Generative Distribution Prediction: A Unified Approach to Multimodal Learning
GDP predicts by sampling candidate outputs from a conditional generative model and selecting the one with the lowest average loss; its excess risk is bounded by generation error plus a vanishing sampling term.