AI Paper Atlas

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

Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos

Yicheng Feng, Wanpeng Zhang, Ye Wang, Hao Luo, Haoqi Yuan, Sipeng Zheng, Zongqing Lu

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 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
remove spatial-aware pretraining (i.e. add it) (ablation)VIPA-VLA · LIBEROavg success (%): 91.2 -> 92.4Improved
add dual 3D encoder (Cut3R) fusion (ablation)VIPA-VLA · LIBEROavg success (%): 90.4 -> 92.4Improved
add both pretraining and dual encoder (ablation)VIPA-VLA · LIBEROavg success (%): 88.7 -> 92.4Improved
pretrain on Hand3D visual annotations (ablation)InternVL3.5 · Hand3D-testdistance error (m): 0.18 -> 0.14Improved
add 3D encoder with fusion on top of Hand3D pretraining (ablation)InternVL3.5 + Hand3D · Hand3D-testdistance error (m): 0.14 -> 0.12Improved
spatial-aware pretraining vs GR00T N1.5 on Wipe-BoardVIPA-VLA · Real robot Wipe-Boardwhole-task success (%): 30% -> 60%Improved
spatial-aware pretraining on Put-Three-Obj whole-task vs GR00T N1.5VIPA-VLA · Real robot Put-Three-Objwhole-task success (%): 40% -> 10%Hurt
VIPA-VLA vs GR00T N1.5 sub-task successVIPA-VLA · Real robot Put-Three-Objsub-task success (%): 48% -> 52%Improved
VIPA-VLA in unseen environments vs baselinesVIPA-VLA · Real robot Wipe-Board-Unseenwhole-task success (%): 10% -> 50%Improved
human-video pretraining without robot data vs π0.5 (two-view)VIPA-VLA · LIBERO avgsuccess rate (%): 96.9 -> 96.8No clear effect
VIPA-VLA single-view vs GR00T N1.5VIPA-VLA · LIBERO avgsuccess rate (%): 92.1 -> 92.4No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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