AI Paper Atlas

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

Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment

Jacky Kwok, Xilun Zhang, Mengdi Xu, Yuejiang Liu, Azalia Mirhoseini, Chelsea Finn, Marco Pavone

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 25 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 CoVer verifier test-time selectionpi0 · SIMPLER in-distribution (red-team instructions)success rate avg (%): 41.5 -> 57.0Improved
add CoVer verifier test-time selectionpi0 · SIMPLER out-of-distribution (red-team instructions)success rate avg (%): 29.7 -> 61.0Improved
add CoVer on top of rephrase-augmented policypi0 (rephrase) · SIMPLER in-distribution (red-team instructions)success rate avg (%): 44.0 -> 65.5Improved
use CoVer vs training-time instruction augmentation (same data)pi0 · SIMPLER out-of-distribution (red-team instructions)success rate avg (%): 48.7 -> 61.0Improved
verify with CoVer vs random rephrase without verification (ablation)pi0 · SIMPLER out-of-distribution (red-team instructions)success rate avg (%): 28.7 -> 61.0Improved
use CoVer instead of RoboMonkey verifierpi0 · SIMPLER in-distribution (red-team instructions)success rate avg (%): 7.5 -> 57.0Improved
add CoVer to pi0 on Zucchini on Towel (OOD)pi0 · SIMPLER Zucchini on Towel (OOD)success rate (%): 30 -> 41Improved
add CoVer to pi0 on Carrot on Platepi0 · SIMPLER Carrot on Plate (in-distribution)success rate (%): 48 -> 48No clear effect
scale sampled action candidates from 1 to 16 with CoVer (ablation)CoVer-VLA · Bridge V2 held-out action erroraction RMSE: 0.166 -> 0.147Improved
use instruction rephrasing instead of repeated sampling or Gaussian perturbation (ablation)action proposal · Bridge V2 test-time scalingoracle action errorImproved
scale rephrases and actions jointly (hybrid sampling) (ablation)action proposal · Bridge V2 test-time scalingoracle action errorImproved
use verification instead of scaling policy training (training compute)CoVer · training costrelative FLOPs: 16.0x -> 3.8xImproved

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