REVIEW 15 cited by
Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
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.
Forward citations
Cited by 15 Pith papers
-
A Sharp KL-Convergence Analysis for Diffusion Models under Minimal Assumptions
A new analysis shows O~(d/epsilon) steps suffice for KL-close diffusion sampling under only L2 score error and finite second moment assumptions, improving the known O~(d/epsilon^2).
-
Spatial Attention: Adapting Execution Horizons for Diffusion Policies via Observation Sensitivity
Under a fixed sampling budget, execution horizons that minimize disturbance-induced likelihood drop should shorten as Spatial Attention rises; forecasting it yields higher success rates than fixed horizons.
-
Amortized Moment Matching for Visual Generation
Amortized Fréchet Distance uses neural nets to match conditional means and covariances, yielding stronger one-step visual generators than explicit FD-loss or multi-step teachers.
-
Flow Map Learning via Nongradient Vector Flow
SGFlow learns the integral map of a probability-flow ODE via a stop-gradient loss whose only stationary point is the true flow map, and it reaches the best-in-comparison FID at 10 steps on CIFAR-10.
-
FlashDecoder: Real-Time Latent-to-Pixel Streaming Decoder with Transformers
A pure-Transformer streaming decoder with a fixed-size rolling KV cache decodes video latents 3.6–4.7x faster than 3D convolutional decoders with comparable PSNR.
-
Dual-End Consistency Model
DE-CM trains a flow-map consistency model on three sub-trajectories (coupling, instantaneous, noise-to-noisy) and reports 1.70 FID one-step on ImageNet 256.
-
Transition Matching Distillation for Fast Video Generation
Splitting a video diffusion model into a fixed feature extractor and a small recurrent flow head lets TMD generate videos in one to two effective steps with better VBench scores than prior distilled models.
-
Transition Models: Rethinking the Generative Learning Objective
TiM trains a single diffusion-type model on arbitrary time-interval transitions, achieving strong one-step and multi-step text-to-image generation with 865M parameters.
-
ShortFT: Diffusion Model Alignment via Shortcut-based Fine-Tuning
ShortFT fine-tunes Stable Diffusion by backpropagating reward gradients through a distilled few-step shortcut denoising chain, improving alignment scores over DRaFT-LV and DRTune.
-
Yume: An Interactive World Generation Model
A diffusion-based video model generates extendable, keyboard-controlled walkthroughs from a single input image, using quantized camera actions as text prompts.
-
Continuous Semi-Implicit Models
CoSIM extends hierarchical semi-implicit variational inference to continuous time, yielding a simulation-free, multistep consistency-style distillation of pretrained diffusion models.
-
Schr\"odinger Bridge Mamba for One-Step Speech Enhancement
A Mamba-based speech enhancer trained with Schrödinger Bridge objectives produces strong denoising and dereverberation in one inference step with a low real-time factor.
-
CHORDS: Diffusion Sampling Accelerator with Multi-core Hierarchical ODE Solvers
CHORDS accelerates diffusion sampling by running hierarchical ODE solvers on multiple cores, with slower solvers rectifying faster ones, achieving up to 2.9x speedup without retraining.
-
Straighten Viscous Rectified Flow via Noise Optimization
VRFNO claims state-of-the-art one- and few-step image generation by straightening rectified flow trajectories, but its sampler relies on real images from the dataset.
-
Reinforcement Learning: From Algorithms To Foundation Models
A dissertation uniting the author's published results: non-exploitable Nash-DQN policies and the FightLadder benchmark for games, plus diffusion/consistency-model world models for RL — a compilation rather than new results.
Discussion (0). Sign in to comment.