APEX derives self-adversarial gradients from condition-shifted velocity fields in flow models to achieve high-fidelity one-step generation, outperforming much larger models and multi-step teachers.
Efficient generative model training via embedded representation warmup.arXiv preprint arXiv:2504.10188
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2roles
background 1polarities
background 1representative citing papers
Data Warmup accelerates diffusion training on ImageNet by scheduling images from low to high complexity via a foreground-based metric and temperature-controlled sampler, improving FID and IS scores faster than uniform sampling.
citing papers explorer
-
Self-Adversarial One Step Generation via Condition Shifting
APEX derives self-adversarial gradients from condition-shifted velocity fields in flow models to achieve high-fidelity one-step generation, outperforming much larger models and multi-step teachers.
-
Data Warmup: Complexity-Aware Curricula for Efficient Diffusion Training
Data Warmup accelerates diffusion training on ImageNet by scheduling images from low to high complexity via a foreground-based metric and temperature-controlled sampler, improving FID and IS scores faster than uniform sampling.