AI Paper Atlas

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

XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representations

Shichao Fan, Kun Wu, Zhengping Che, Xinhua Wang, Di Wu, Fei Liao, Ning Liu, Yixue Zhang, Zhen Zhao, Zhiyuan Xu, Meng Li, Qingjie Liu and 3 more

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 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 UVMC-guided pretraining (Stages 1-2) to lightweight model (ablation)XR-1-Light · Dual-Arm UR-5e, 6 tasksaverage success rate (%): 42.5 -> 57.5Improved
add cross-modal KL alignment loss between vision and motion codes (ablation)XR-1 · Dual-Arm UR-5e, 6 tasksaverage success rate (%): 48.3 -> 66.7Improved
add UVMC learning on downstream data (no large-scale pretraining) (ablation)XR-1 · Dual-Arm UR-5e, 6 tasksaverage success rate (%): 28.3 -> 48.3 (w/o KL) -> 66.7Improved
scale Stage-1 pretraining data from 1% to 100% (ablation)XR-1 · Dual-Arm UR-5e, 6 tasks, no fine-tuningaverage success rate (%): 29.2 -> 38.3 -> 53.3 -> 65.0Improved
pretrain Stage-2 on XR-D cross-embodiment data and fine-tune, versus downstream-only Stage-1/2 (ablation)XR-1 · Dual-Arm UR-5e, 6 tasksaverage success rate (%): 66.7 -> 81.6Improved
use XR-1 instead of pi0 on unseen scenariosXR-1 · Dual-Arm Franka SweepTrash, dynamic distractorssuccess rate (%): 5 -> 55Improved
use XR-1 instead of baselines on unseen embodiment Tien Kung 2.0XR-1 · Tien Kung 2.0, 20 tasksaverage success rate (%): 41.0 (pi0.5) -> 72.0Improved

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