AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation

Hansheng Chen, Kai Zhang, Hao Tan, Leonidas Guibas, Gordon Wetzstein, Sai Bi

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Rectified flow straightening · 25 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use GM policy instead of DX policy (ablation)pi-Flow (REPA teacher) · ImageNet 256 (1-NFE), 40K iterationsFID: 4.44 -> 3.07Improved
add GM dropout 0.05 (ablation)pi-Flow GM K=32 (FM teacher) · ImageNet 256 (1-NFE)FID: 4.14 -> 3.65Improved
increase GM components K from 8 to 32 (ablation)pi-Flow GM (REPA teacher) · ImageNet 256 (1-NFE), 40K iterationsFID: 3.07 -> 3.08No clear effect
vary DX grid points N from 20 to 40 (ablation)pi-Flow DX policy · ImageNet 256 (1-NFE), 40K iterationsFID: 4.44 -> 4.90Hurt
use REPA teacher instead of FM teacherpi-Flow GM K=32 · ImageNet 256 (1-NFE)FID: 3.65 -> 3.08Improved
add GM temperature (ablation)pi-Flow (GM-FLUX) 4-NFE · HPSv2 prompts, teacher-referencedFID: 14.9 -> 14.3Improved
add micro-window velocity matching (ablation)pi-Flow (GM-FLUX) 4-NFE · HPSv2 prompts, teacher-referencedFID: 14.6 -> 14.3Improved
use pi-ID instead of VSD distillation for diversityQwen-Image Lightning · OneIG-Bench (Qwen-Image, 4-NFE)Diversity: 0.116 -> 0.180Improved
use pi-ID instead of VSD distillation for diversitySenseFlow (FLUX) · OneIG-Bench (FLUX, 4-NFE)Diversity: 0.151 -> 0.229Improved
use pi-ID instead of VSD for preferenceQwen-Image Lightning · COCO-10k (Qwen-Image, 4-NFE)HPSv2.1: 0.322 -> 0.314Hurt
use pi-ID instead of VSD for text renderingQwen-Image Lightning · OneIG-Bench (Qwen-Image, 4-NFE)Text: 0.923 -> 0.892Hurt
use data-free instead of data-dependent pi-IDpi-Flow (GM-Qwen) · COCO-10k (Qwen-Image, 4-NFE)FID: 36.0 -> 36.0No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy