A training-free predictor-corrector method that accelerates Diffusion Transformers by solving a feature-ODE, achieving large compute reductions with modest quality loss.
In Ad- vances in Neural Information Processing Systems 30 (NIPS 2017)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Forecast then Calibrate: Feature Caching as ODE for Efficient Diffusion Transformers
A training-free predictor-corrector method that accelerates Diffusion Transformers by solving a feature-ODE, achieving large compute reductions with modest quality loss.