Skip to content
RobotWorld
← Back to Videos

DIDO: One-Step Denoising for World Action Models

Loading video
Loading video

DIDO distills a four-step video world model into one denoising step, preserving gripper-object dynamics for manipulation while cutting action latency from 562 ms to 384 ms.

World-action modelsRobot ManipulationVideo GenerationModel DistillationRobotic ArmBAAIGalbot
Category: arm
Author: @stepjamUK
Date: 2026-09-15T00:00:00
Duration: 68.091s
Source: https://x.com/stepjamUK/status/2099860791008674014
Reference: https://arxiv.org/abs/2609.15570

Related Videos

OneWM: World and Action Models That Strengthen Each Other
2:52

OneWM: World and Action Models That Strengthen Each Other

@Selen7005717917 · 119 views · 2026-10-24
World ModelsRobot Manipulation
3DPWM: Completed Point Clouds for Robot Planning
0:55

3DPWM: Completed Point Clouds for Robot Planning

@fuxinli2 · 153 views · 2026-10-20
3DPWMPoint Cloud Completion
R2-WAM: data-free continual learning for world action models via rectified generative replay
1:14

R2-WAM: data-free continual learning for world action models via rectified generative replay

@Yunsong_Zhou · 144 views · 2026-10-21
ContinualLearningWorldActionModel
Maker Mods Arms Run Plain-Language Tasks on Dream-1
1:02

Maker Mods Arms Run Plain-Language Tasks on Dream-1

@IsaacSin12 · 140 views · 2026-10-19
Robotic ArmVLA
ASPIRE: Model Frontier Meets Robot Cost
1:18

ASPIRE: Model Frontier Meets Robot Cost

@DrJimFan · 196 views · 2026-10-16
NVIDIAASPIRE
ASPIRE: Open-Source Robotic Skill Discovery
1:18

ASPIRE: Open-Source Robotic Skill Discovery

@guanzhi_wang · 176 views · 2026-10-16
NVIDIAASPIRE

As an Amazon Associate, we earn from qualifying purchases.