MCLR: Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives.

arXiv:2603.22364v4 Announce Type: replace-cross Abstract: Diffusion models achieve strong generative performance but often rely on classifier-free guidance (CFG), an inference-time intervention that modifies the sampling trajectory. Since standard denoising sc