AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

RehearseVLA: Simulated Post-Training for VLAs with Physically-Consistent World Model

Junjin Xiao, Yandan Yang, Xinyuan Chang, Ronghan Chen, Feng Xiong, Mu Xu, Wei-Shi Zheng, Qing Zhang

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models

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 world-model RL post-training (RehearseVLA)OpenVLA-OFT · LIBERO average (5 demos/task)success rate (%): 74.85 -> 79.6Improved
add RL post-training in world modelOpenVLA-OFT · LIBERO-Long (5 demos/task)success rate (%): 57.0 -> 57.8No clear effect
add RL post-training in world modelOpenVLA-OFT · LIBERO-Object (5 demos/task)success rate (%): 74.2 -> 86.6Improved
add extra exploration data to world simulator (ablation)OpenVLA-OFT + RehearseVLA · LIBERO-Goal (5 demos/task)success rate (%): 68.4 -> 79.8Improved
add trained reward head (vs prompted VLM) (ablation)OpenVLA-OFT + RehearseVLA · LIBERO-Goal (5 demos/task)success rate (%): 68.4 -> 68.8No clear effect
add both extra data and reward head (ablation)OpenVLA-OFT + RehearseVLA · LIBERO-Goal (5 demos/task)success rate (%): 68.4 -> 86.4Improved
add both extra data and reward head (ablation)OpenVLA-OFT + RehearseVLA · LIBERO-Long (5 demos/task)success rate (%): 42.2 -> 57.8Improved
add reflector-based terminationOpenVLA-OFT + RehearseVLA · LIBERO average, no GT terminationsuccess rate (%): 63.05 -> 74.9Improved
use RehearseVLA instead of simulator-based RIPT-VLAOpenVLA-OFT · LIBERO suites (5 demos/task)success rate (%): Object 83.4 -> 86.6; Spatial 88.6 -> 87.6Mixed
apply RehearseVLA post-training in real worldOpenVLA-OFT · Real-world taskssuccess rate (%): Clean table 20 -> 30; green toy 30 -> 50Improved

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