Establishes a formal mathematical equivalence between Classifier-Free Guidance (CFG) and alignment-based objectives, allowing for CFG-like quality without inference-time overhead.
March 25, 2026
Original Paper
MCLR: Improving Conditional Modeling in Visual Generative Models via Inter-Class Likelihood-Ratio Maximization and Establishing the Equivalence between Classifier-Free Guidance and Alignment Objectives
arXiv · 2603.22364
The Takeaway
By reframing CFG as an inter-class likelihood-ratio maximization during training, practitioners can train models that possess the 'guidance' effect inherently. This eliminates the need for the double-pass sampling required by traditional CFG, potentially halving generation latency.
From the abstract
Diffusion models have achieved state-of-the-art performance in generative modeling, but their success often relies heavily on classifier-free guidance (CFG), an inference-time heuristic that modifies the sampling trajectory. From a theoretical perspective, diffusion models trained with standard denoising score matching (DSM) are expected to recover the target data distribution, raising the question of why inference-time guidance is necessary in practice. In this work, we ask whether the DSM trai