Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion

Dohoon Lee, Jaehyun Park, Hyunwoo J. Kim, Kyogu Lee
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:33635-33654, 2025.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-lee25ac, title = {Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion}, author = {Lee, Dohoon and Park, Jaehyun and Kim, Hyunwoo J. and Lee, Kyogu}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {33635--33654}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/lee25ac/lee25ac.pdf}, url = {https://proceedings.mlr.press/v267/lee25ac.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion %A Dohoon Lee %A Jaehyun Park %A Hyunwoo J. Kim %A Kyogu Lee %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-lee25ac %I PMLR %P 33635--33654 %U https://proceedings.mlr.press/v267/lee25ac.html %V 267 %X 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.
APA
Lee, D., Park, J., Kim, H.J. & Lee, K.. (2025). Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:33635-33654 Available from https://proceedings.mlr.press/v267/lee25ac.html.

Related Material