Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
read the original abstract
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.
This paper has not been read by Pith yet.
Forward citations
Cited by 27 Pith papers
-
How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance
FMRG reformulates guidance as deterministic optimal control, deriving a single-trajectory method using the flow map that matches or exceeds baselines on reward-guided generation and inverse problems with 3 NFEs at tex...
-
Multimarginal flow matching with optimal transport potentials
OTP-FM extends conditional flow matching by incorporating dynamic optimal transport potentials to enable efficient multimarginal transport learning with intermediate observed marginals.
-
How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance
FMRG is a training-free, single-trajectory guidance method for flow models derived from optimal control that achieves strong reward alignment with only 3 NFEs.
-
Causal-rCM: A Unified Teacher-Forcing and Self-Forcing Open Recipe for Autoregressive Diffusion Distillation in Streaming Video Generation and Interactive World Models
Causal-rCM unifies teacher-forcing and self-forcing distillation for autoregressive video diffusion, delivering a 2-step model with VBench-T2V score 84.63 and enabling interactive world models on Cosmos 3 using only s...
-
DiffusionBench: On Holistic Evaluation of Diffusion Transformers
NanoGen unifies DiT training on ImageNet and T2I, reveals negative Pearson correlations (-0.377 to -0.580) in method rankings across metrics from 21 models, and motivates DiffusionBench for holistic evaluation.
-
Few-step Generative Models as Lossy Compression
Few-step generative models can be reformulated as lossy codecs in the reverse channel coding framework without retraining, yielding faster encoding/decoding on low-resolution image benchmarks.
-
Few-step Cofolding with All-Atom Flow Maps
DeCAF distills all-atom cofolding diffusion models into few-step flow maps, showing improved or matched accuracy on protein-ligand tasks with 5x fewer inference steps.
-
LIFT and PLACE: A Simple, Stable, and Effective Knowledge Distillation Framework for Lightweight Diffusion Models
LIFT decomposes distillation into coarse linear alignment then fine refinement while PLACE adds error-based local adaptation, allowing stable training of 1.3M-parameter students (1.6% teacher size) to FID 15.73 across...
-
LIFT and PLACE: A Simple, Stable, and Effective Knowledge Distillation Framework for Lightweight Diffusion Models
LIFT and PLACE enable stable knowledge distillation for extremely lightweight diffusion models by decomposing the task into coarse alignment followed by fine refinement with piecewise local adaptive guidance.
-
LIFT and PLACE: A Simple, Stable, and Effective Knowledge Distillation Framework for Lightweight Diffusion Models
LIFT and PLACE enable stable training of extremely compressed diffusion models by breaking distillation into coarse linear alignment followed by local adaptive refinement.
-
Dual-Rate Diffusion: Accelerating diffusion models with an interleaved heavy-light network
Dual-Rate Diffusion interleaves sparse heavy context encoding with frequent light denoising to cut diffusion sampling cost by 2-4x on ImageNet while matching baseline quality and remaining compatible with distillation.
-
Thermal-Only Crowd Counting with Deployment-Time Privacy Protection
A privacy-preserving thermal-only crowd counting framework extracts enhanced features from thermal images via single-step LCM denoising in a depth-to-RGB diffusion model and matches RGB-T fusion performance without RG...
-
Physical Fidelity Reconstruction via Improved Consistency-Distilled Flow Matching for Dynamical Systems
Distilled one-step consistency model from optimal-transport flow-matching teacher reconstructs high-fidelity dynamical system flows from low-fidelity data with 12x speedup, half the parameters, and 23.1% better SSIM t...
-
How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance
FMRG is a training-free single-trajectory guidance framework for flow-based models that matches or exceeds baselines on reward-guided tasks and inverse problems using as few as 3 NFEs.
-
Efficient Diffusion Distillation via Embedding Loss
Embedding Loss aligns feature distributions via MMD in random network embeddings to boost one-step diffusion distillation, reaching SOTA FID of 1.475 on CIFAR-10 unconditional generation.
-
Allo{SR}$^2$: Rectifying One-Step Super-Resolution to Stay Real via Allomorphic Generative Flows
Allo{SR}^2 rectifies one-step super-resolution trajectories with allomorphic generative flows via SNR initialization, velocity supervision, and self-adversarial matching to deliver state-of-the-art fidelity and realism.
-
Post-Hoc Guidance for Consistency Models by Joint Flow Distribution Learning
JFDL allows pre-trained Consistency Models to perform guided image generation post-hoc by aligning flow distributions, reducing FID scores on CIFAR-10 and ImageNet without needing a teacher model.
-
Dual-End Consistency Model
DE-CM reaches state-of-the-art one-step FID of 1.70 on ImageNet 256x256 by decomposing PF-ODE trajectories into three critical sub-trajectories and using flow matching plus N2N mapping for stability.
-
Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency
The work introduces rCM, a score-regularized continuous-time consistency model that matches DMD2 quality on large models up to 14B parameters while improving diversity and enabling 1-4 step sampling.
-
DOLLAR: Few-Step Video Generation via Distillation and Latent Reward Optimization
DOLLAR combines variational score and consistency distillation for few-step video generation plus latent reward optimization, reporting 82.57 VBench score and up to 278x speedup over the teacher diffusion model for 12...
-
Improved DDIM Sampling with Moment Matching Gaussian Mixtures
Moment-matched GMM kernels in DDIM yield lower FID and higher IS than Gaussian kernels at small sampling steps on CelebA-HQ, FFHQ, ImageNet, and Stable Diffusion tasks.
-
Data-Forcing Distillation: Restoring Diversity and Fidelity in Few-Step Video Generation
Data-Forcing Distillation adds a teacher score discrepancy term to DMD-style distillation, restoring diversity and fidelity in few-step video models with 100-300 finetuning steps.
-
Trajectory-Consistent Flow Matching for Robust Visuomotor Policy Learning
Trajectory consistency training, smoothness regularization, and higher-order integration for flow matching policies deliver 60-70% success on long-horizon real-robot tasks where baselines achieve 0%.
-
Consistency Regularised Gradient Flows for Inverse Problems
A consistency-regularized Euclidean-Wasserstein-2 gradient flow performs joint posterior sampling and prompt optimization in latent space for efficient low-NFE inverse problem solving with diffusion models.
-
Teacher-Feature Drifting: One-Step Diffusion Distillation with Pretrained Diffusion Representations
A simplified one-step diffusion distillation uses pretrained teacher features directly for drifting loss plus a mode coverage term, achieving FID 1.58 on ImageNet-64 and 18.4 on SDXL.
-
Stabilizing, Scaling & Enhancing MeanFlow for Large-scale Diffusion Distillation
Stabilizes MeanFlow for large-scale diffusion distillation via discrete warm-up and trajectory alignment, reporting better results on FLUX.1-dev and HunyuanImage 3.0.
-
A Unified Measure-Theoretic View of Diffusion, Score-Based, and Flow Matching Generative Models
Diffusion, score-based, and flow matching models are unified as instances of learning time-dependent vector fields inducing marginal distributions governed by continuity and Fokker-Planck equations.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.