Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
By Dongjun Kim, Chieh-Hsin Lai, Wei-Hsiang Liao, Naoki Murata, Yuhta Takida, Toshimitsu Uesaka, Yutong He, Yuki Mitsufuji, Stefano Ermon. International Conference on Learning Representations, 2023
438 citations recorded by the source corpus.
Consistency Models (CM) (Song et al., 2023) accelerate score-based diffusion model sampling at the cost of sample quality but lack a natural way to trade-off quality for speed. To address this limitation, we propose Consistency Trajectory Model (CTM), a generalization encompassing CM and score-based models as special cases. CTM trains a single neural network that can -- in a single forward pass -- output scores (i.e., gradients of log-density) and enables unrestricted traversal between any initial and final time along the Probability Flow Ordinary Differential Equation (ODE) in a diffusion process. CTM enables the efficient combination of adversarial training and denoising score matching loss to enhance performance and achieves new state-of-the-art FIDs for single-step diffusion model sampling on CIFAR-10 (FID 1.73) and ImageNet at 64x64 resolution (FID 1.92). CTM also enables a new family of sampling schemes, both deterministic and stochastic, involving long jumps along the ODE solution trajectories. It consistently improves sample quality as computational budgets increase, avoiding the degradation seen in CM. Furthermore, unlike CM, CTM's access to the score function can streamline the adoption of established controllable/conditional generation methods from the diffusion community. This access also enables the computation of likelihood. The code is available at https://github.com/sony/ctm.
Subfields: Artificial Intelligence
View Paper: https://www.semanticscholar.org/paper/2ea9167ca0dd536923ea0968d6b89e2b6cc34f21
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Richard Zhang
- Eli Shechtman
- Andreas Blattmann
- Frédo Durand
- Robin Rombach
- William T. Freeman
- Zico Kolter
- Dominik Lorenz
Citations in the corpus
- Mean Flows for One-step Generative Modeling (2025, positive): “While related to previous one-step generative models [46, 43, 14, 31, 49, 23, 12, 52], our method provides a more principled framework.”
- Improved Distribution Matching Distillation for Fast Image Synthesis (2024, methodology): “Consistency Distillation [9,12,19,26,31,32], and TRACT [33], train student models so their outputs are self-consistent at any timesteps along the ODE trajectory, and thus consistent with the teacher.”
- Adversarial Diffusion Distillation (2023, methodology): “To improve quality, a discriminator loss is added to the score matching objective in Adversarial Score Matching [24] and the consistency objective of CTM [29].”
- From Slow Bidirectional to Fast Autoregressive Video Diffusion Models (2024, positive): “Unlike trajectory-preserving methods [32, 47, 69], DMD provides supervision at the distribution level and offers the unique advantage of allowing different architectural formulations for the teacher and student diffusion models.”