Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion
This work addresses a specific problem in generative modeling for researchers and practitioners, offering an incremental improvement through a novel plug-in module.
The paper tackles the limitations of flow and diffusion models in terms of dimensionality freedom and adaptability to inference trajectories by proposing the Multidimensional Adaptive Coefficient (MAC), a plug-in module that extends unidimensional coefficients to multidimensional ones and enables trajectory-wise adaptation. Empirical results show MAC enhances generative quality with high training efficiency across diverse frameworks and datasets.
Flow and diffusion models have demonstrated strong performance and training stability across various tasks but lack two critical properties of simulation-based methods: freedom of dimensionality and adaptability to different inference trajectories. To address this limitation, we propose the Multidimensional Adaptive Coefficient (MAC), a plug-in module for flow and diffusion models that extends conventional unidimensional coefficients to multidimensional ones and enables inference trajectory-wise adaptation. MAC is trained via simulation-based feedback through adversarial refinement. Empirical results across diverse frameworks and datasets demonstrate that MAC enhances generative quality with high training efficiency. Consequently, our work offers a new perspective on inference trajectory optimality, encouraging future research to move beyond vector field design and to leverage training-efficient, simulation-based optimization.