AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

MoMaGen: Generating Demonstrations under Soft and Hard Constraints for Multi-Step Bimanual Mobile Manipulation

Chengshu Li, Mengdi Xu, Arpit Bahety, Hang Yin, Yunfan Jiang, Huang Huang, Josiah Wong, Sujay Garlanka, Cem Gokmen, Ruohan Zhang, Weiyu Liu, Jiajun Wu and 2 more

ICLR 2026 · Reinforcement learning › Visual and representation RL · approach: Dexterous manipulation transfer · 14 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 hard and soft visibility constraints (ablation)MoMaGen · Tidy Table D0 object visibilityvisibility ratio: 0.46 -> 0.86Improved
add hard and soft visibility constraints (ablation)MoMaGen · Clean Frying Pan D1 object visibilityvisibility ratio: 0.43 -> 0.80Improved
add hard and soft visibility constraints (ablation)MoMaGen · Tidy Table D0 data generationsuccess rate: 0.74 -> 0.80Improved
add visibility constraints (ablation)MoMaGen · Pick Cup D0 data generationsuccess rate: 0.97 -> 0.86Hurt
add hard visibility constraint (ablation)MoMaGen w/o hard · Clean Frying Pan D0 data generationsuccess rate: 0.24 -> 0.51Improved
add soft visibility constraint (ablation)MoMaGen w/o soft · Put Dishes Away D0 data generationsuccess rate: 0.50 -> 0.38Hurt
use MoMaGen instead of SkillMimicGen with base replaySkillMimicGen · Tidy Table D0 object visibilityvisibility ratio: 0.40 -> 0.86Improved
use MoMaGen instead of SkillMimicGenSkillMimicGen · Pick Cup D0 data generationsuccess rate: 1.00 -> 0.86Hurt
add visibility constraints to data generation (ablation)WB-VIMA policy · Tidy Table D0 policy successsuccess rate: 0.05 -> 0.40Improved
add visibility constraints to data generation (ablation)WB-VIMA policy · Pick Cup D0 policy successsuccess rate: 0.65 -> 0.75Improved
train on MoMaGen D0+ generated data instead of D0 baseline dataWB-VIMA policy · Pick Cup D1 policy successsuccess rate: 0 -> 0.25Improved
pretrain on MoMaGen sim data before real fine-tunepi0 · Real-world Pick Cupsuccess rate (%): 0 -> 60Improved

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