Using understanding tasks as direct supervision during post-training improves image generation and editing in unified multimodal models.
Eq-vae: Equivariance regularized latent space for improved generative image modeling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Steering Visual Generation in Unified Multimodal Models with Understanding Supervision
Using understanding tasks as direct supervision during post-training improves image generation and editing in unified multimodal models.