CT-Physics trains a consistency model in two stages, first on data alone and then with a physical constraint loss, to generate constraint-satisfying samples in one step.
Scaling learning algorithms towards AI
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Consistency Training with Physical Constraints
CT-Physics trains a consistency model in two stages, first on data alone and then with a physical constraint loss, to generate constraint-satisfying samples in one step.