AI Paper Atlas

Research map / Video / Video segmentation and tracking

Predicting What Matters: Robust Generalist Robot Policy Learning via Future Semantic Mask

Yunfan Lou, Xiaowei Chi, Xiaojie Zhang, Zezhong Qian, Chengxuan Li, Rongyu Zhang, Yaoxu Lyu, Guoyu Song, Chuyao Fu, Haoxuan Xu, Pengwei Wang, Shanghang Zhang

ICML 2026 · Video › Video segmentation and tracking · approach: Stochastic video prediction

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
predict future semantic masks (MWM-C1) instead of RGB (Cosmos w/ IDM)IDM-based world model · LIBERO averagesuccess rate: 0.675 -> 0.810Improved
predict future semantic masks (MWM-C1) instead of RGB (Cosmos w/ IDM)IDM-based world model · LIBERO-10success rate: 0.488 -> 0.704Improved
predict mask latent features (MWM-C2) instead of RGB (Cosmos w/ Latent IDM)latent IDM world model · LIBERO averagesuccess rate: 0.873 -> 0.918Improved
use predictive latent features with an action head instead of explicit decoded masks plus IDM (ablation)MWM · LIBERO averagesuccess rate: 0.810 -> 0.918Improved
add wrist-view input to the mask world model (ablation)MWM-C2 · LIBERO averagesuccess rate: 0.918 -> 0.983Improved
replace GE-ACT RGB world model with MWMGE-ACT · LIBERO averagesuccess rate: 0.965 -> 0.983Improved
replace GE-ACT RGB world model with MWMGE-ACT · RLBench (6 tasks)success rate: 30.8% -> 68.3%Improved
replace FiS-VLA with MWMFiS-VLA · RLBench (6 tasks)success rate: 50.0% -> 68.3%Improved
replace GE-ACT with MWMGE-ACT · Real-world Frankasuccess rate: 23.8% -> 67.5%Improved
replace pi0 with MWMpi0 · Real-world Frankasuccess rate: 38.8% -> 67.5%Improved
replace GE-ACT with MWM under appearance shiftsGE-ACT · Real-world OODOOD-SR: 12.5% -> 42.1%Improved
replace GE-ACT with MWM under background shiftGE-ACT · Real-world background shiftsuccess rate: 3.8% -> 27.5%Improved

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