Decision trees and diffusion models are unified via a shared Global Trajectory Score Matching principle, yielding TreeFlow for tabular generation and DSMTree for distilling tree logic into networks.
This network uses an embedding layer for the tree level j (embedding dim=32) and a 2-hidden-layer MLP (256 units each, with ReLU and BatchNorm) to predict the split decision
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Trees to Flows and Back: Unifying Decision Trees and Diffusion Models
Decision trees and diffusion models are unified via a shared Global Trajectory Score Matching principle, yielding TreeFlow for tabular generation and DSMTree for distilling tree logic into networks.