Pith. sign in

REVIEW 10 cited by

Transition Models: Rethinking the Generative Learning Objective

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

arxiv 2509.04394 v1 pith:SI2PQHNJ submitted 2025-09-04 cs.LG cs.CV

Transition Models: Rethinking the Generative Learning Objective

classification cs.LG cs.CV
keywords generativemodelsparametersqualityacrossdynamicsfew-stepfidelity
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

A fundamental dilemma in generative modeling persists: iterative diffusion models achieve outstanding fidelity, but at a significant computational cost, while efficient few-step alternatives are constrained by a hard quality ceiling. This conflict between generation steps and output quality arises from restrictive training objectives that focus exclusively on either infinitesimal dynamics (PF-ODEs) or direct endpoint prediction. We address this challenge by introducing an exact, continuous-time dynamics equation that analytically defines state transitions across any finite time interval. This leads to a novel generative paradigm, Transition Models (TiM), which adapt to arbitrary-step transitions, seamlessly traversing the generative trajectory from single leaps to fine-grained refinement with more steps. Despite having only 865M parameters, TiM achieves state-of-the-art performance, surpassing leading models such as SD3.5 (8B parameters) and FLUX.1 (12B parameters) across all evaluated step counts. Importantly, unlike previous few-step generators, TiM demonstrates monotonic quality improvement as the sampling budget increases. Additionally, when employing our native-resolution strategy, TiM delivers exceptional fidelity at resolutions up to 4096x4096.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation

    cs.CV 2026-05 unverdicted novelty 8.0

    AnyFlow enables any-step video diffusion by distilling flow-map transitions over arbitrary time intervals with on-policy backward simulation.

  2. One-Step Generative Modeling via Wasserstein Gradient Flows

    cs.LG 2026-05 conditional novelty 7.0

    W-Flow achieves state-of-the-art one-step ImageNet 256x256 generation at 1.29 FID by training a static neural network to follow a Wasserstein gradient flow that minimizes Sinkhorn divergence, delivering roughly 100x f...

  3. Parallel Decoding Distillation for Fast Image and Video Generation

    cs.CV 2026-07 conditional novelty 6.0

    A trajectory-based distillation method trains a student to predict multiple mean velocities per network evaluation, enabling 4-8 step generation with competitive quality and improved diversity.

  4. One-Step Generative Modeling via Wasserstein Gradient Flows

    cs.LG 2026-05 unverdicted novelty 6.0

    W-Flow compresses a Wasserstein gradient flow defined via Sinkhorn divergence into a single-step neural generator, reporting 1.29 FID on ImageNet 256x256 with improved mode coverage.

  5. Extending One-Step Image Generation from Class Labels to Text via Discriminative Text Representation

    cs.CV 2026-04 unverdicted novelty 6.0

    By requiring and using highly discriminative LLM text features, the work enables the first effective one-step text-conditioned image generation with MeanFlow.

  6. Transition Matching Distillation for Fast Video Generation

    cs.CV 2026-01 conditional novelty 6.0

    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.

  7. Understanding, Accelerating, and Improving MeanFlow Training

    cs.CV 2025-11 conditional novelty 6.0

    Training MeanFlow by first forming instantaneous velocity and short-gap average velocity, then shifting to long gaps, improves 1-NFE ImageNet FID from 3.43 to 2.87 and speeds training by about 2.5x.

  8. Trajectory Forcing: Structure-First Generation with Controllable Semantic Trajectories

    cs.CV 2026-06 unverdicted novelty 5.0

    Trajectory Forcing makes generative image synthesis trajectory-centric by organizing it into decodable semantic stages derived from clustered visual representations and trained with one-step flow-matching models.

  9. SubFlow: Sub-mode Conditioned Flow Matching for Diverse One-Step Generation

    cs.LG 2026-04 unverdicted novelty 5.0

    SubFlow restores full mode coverage in one-step flow matching by conditioning on sub-modes from semantic clustering, yielding higher diversity on ImageNet-256 while preserving FID.

  10. Improved Mean Flows: On the Challenges of Fastforward Generative Models

    cs.CV 2025-12 unverdicted novelty 5.0

    Improved MeanFlow (iMF) reaches 1.72 FID on ImageNet 256x256 with one function evaluation by reformulating the training objective as a regression on instantaneous velocity and treating guidance as flexible conditionin...