arXiv Computer Vision October 7, 2026 By Zhen Guo, Rongyuan Wu, Qiaosi Yi, Chenxi Xie, Xinyu Wei, Lei Zhang

Two Halves are More than One: Phase-wise Velocity Distillation for Fast and High-Quality Image Generation

Read the original on arXiv Computer Vision →

The paper introduces Phase-wise Velocity Distillation (PVD), a method that splits the diffusion generation timeline into coarse and fine phases, each handled by a half-sized expert. By decoupling structural composition from detail refinement, PVD achieves higher quality image generation within a single full-backbone compute budget, outperforming monolithic student models. Experiments on ImageNet and text-to-image tasks show competitive FID scores and significant reductions in parameters and VRAM usage.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv Computer Vision

4d ago

Data-free Multistep Diffusion Distillation

arXiv:2610.19305v1 Announce Type: new Abstract: One-step diffusion distillation achieves strong performance on small datasets but remains insufficient for high-resolution or complex text-to-image set...

By Ngoc Duy Tran, Kien Do, Duc Thanh Nguyen, Truyen Tran

diffusion efficiency safety

arXiv AI

Jul 9

Dynamic-in-Few-Step: Unifying Dynamic Computation and Few-Step Distillation for Efficient Video Generation

arXiv:2607. 06631v1 Announce Type: cross Abstract: Video Diffusion Models (VDMs) have demonstrated superior generation quality but suffer from prohibitive computational costs.

By Yu Cheng, Siyue Yao, Zhongang Qi, Shanyan Guan, Wei Li, Fajie Yuan

diffusion efficiency

arXiv Machine Learning

Oct 9

Omni-Diffusion-Distill: Few-Step Distillation of Unified Multimodal Diffusion Large Language Models

Omni‑Diffusion‑Distill is a two‑stage distillation framework that compresses unified multimodal diffusion large language models (dLLMs) into a single efficient student while preserving both image generation and multimodal understanding. In the first stage the student learns to skip decoding steps by replaying cached teacher trajectories, and in the second stage it is refined on intermediate states along its own rollouts. The method introduces a pairwise collision penalty to reduce repetition in parallel text decoding and entropy‑matched guidance to prevent entropy collapse in image generation, achieving state‑of‑the‑art trade‑offs: reducing image generation from 128 to 8 steps and multimodal understanding from 512 to 64 steps, yielding 18.2× and 21.2× wall‑clock speedups and strong performance on GenEval, DPG‑Bench, MM‑Vet, and COCO captioning.

By Hong Huang, Chenhongyi Yang, Junzhe Sun, Animesh Sinha, Wuyang Chen, Yifan Jiang

llms diffusion efficiency multimodal benchmarks

arXiv AI

Jul 13

Transition Matching Distillation for Fast Video Generation

arXiv:2601. 09881v2 Announce Type: replace-cross Abstract: Large video diffusion and flow models have achieved remarkable success in high-quality video generation, but their use in real-time interactive applications remains limited due to their inefficient multi-step sampling process.

By Weili Nie, Julius Berner, Nanye Ma, Chao Liu, Saining Xie, Arash Vahdat

diffusion efficiency

arXiv Computer Vision

Sep 1

Improving Image-to-Image Translation via a Rectified Flow Reformulation

arXiv:2603.20186v2 Announce Type: replace Abstract: In this work, we propose Image-to-Image Rectified Flow Reformulation (I2I-RFR), a practical plug-in reformulation that recasts standard I2I regress...

By Satoshi Iizuka, Shun Okamoto, Kazuhiro Fukui

efficiency multimodal

arXiv Machine Learning

Sep 30

Improved Distributional Diffusion Models

arXiv:2609.37147v1 Announce Type: cross Abstract: Distributional Diffusion Models (DDMs) replace the standard mean-prediction denoiser with a \emph{distributional} denoiser trained via a scoring rule...

By Tommaso Martorella, Alexandre Galashov, Felix Krause, Stefan Andreas Baumann, Valentin De Bortoli, Arthur Gretton, Bj\"orn Ommer

llms diffusion efficiency