AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

STEP: Warm-Started Visuomotor Policies with Spatiotemporal Consistency Prediction

Jinhao Li, Yuxuan Cong, Yingqiao Wang, Hao Xia, Shan Huang, Yijia Zhang, Ningyi Xu, Guohao Dai

ICML 2026 · Image generation › Fast diffusion sampling · approach: Accelerated diffusion policies · 0 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
add velocity-aware perturbation injection (sigma_stall 0 -> 1.2) to STEP (ablation)STEP, 2 steps · Real-world PickNPlacesuccess rate: 0.00 -> 0.95Improved
add velocity-aware perturbation injection (sigma_stall 0 -> 1.2) to STEP (ablation)STEP, 2 steps · Real-world StackCubesuccess rate: 0.00 -> 0.80Improved
increase perturbation scale sigma_stall to 2.0 (ablation)STEP · Real-world PickNPlace and StackCubesuccess rate: 1.00 -> 0.00 (step=8, sigma_stall 1.2 -> 2.0)Hurt
use moderate predictor weight sigma_scale 0.2 or 0.4 instead of 0 (ablation)STEP, 2 steps, sigma_stall=1.2 · Real-world PickNPlacesuccess rate: 0.65 -> 0.95Improved
increase predictor weight sigma_scale from 0.2 to 1.0 (ablation)STEP, 2 steps, sigma_stall=1.2 · Real-world PickNPlacesuccess rate: 0.95 -> 0.65Hurt
use STEP warm-start instead of DDIM at 2 stepsDiffusion policy · Real-world PickNPlacesuccess rate: 0.60 -> 0.95Improved
use STEP warm-start instead of DDIM at 4 stepsDiffusion policy · Real-world StackCubesuccess rate: 0.80 -> 1.00Improved
use STEP instead of DDIM at 2 steps (state-based)Diffusion policy · Transport (state-based)score: 0.04 -> 0.86Improved
use STEP instead of BRIDGER at 2 steps (state-based)Diffusion policy · ToolHang (state-based)score: 0.08 -> 0.64Improved
use STEP instead of BRIDGER at 2 steps (ManiSkill2 PushChair)Diffusion policy · ManiSkill2 PushChairscore: 0.45 -> 0.39Hurt
use STEP instead of BRIDGER at 4 steps (image-based Transport)Diffusion policy · Transport (image-based)score: 0.88 -> 0.88No clear effect
reduce denoising steps from 4 to 2 in STEP (state-based Push-T)STEP · Push-T (state-based)score: 0.91 -> 0.49Hurt

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