AI Paper Atlas

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

VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Model

Yanjiang Guo, Tony Lee, Lucy Xiaoyang Shi, Jianyu Chen, Percy Liang, Chelsea Finn

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
finetune the world model on expert rollouts and then online policy rollouts (ablation)Ctrl-World · action replay, 256 clipsFVD: 225.13 -> 99.98 -> 64.12Improved
add online rollout data (success and failure) to world model finetuning (ablation)Ctrl-World + expert rollout · interaction event confusion matrix, 50 clipsfalse positives: 11 -> 1Improved
add online rollout data to world model finetuning (ablation)Ctrl-World + expert rollout · interaction event confusion matrix, 50 clipstrue positives: 28 -> 26 TP, 9 -> 19 TNMixed
finetune world model on expert demonstrations only (ablation)Ctrl-World · action replay, qualitativeinteraction outcome fidelityHurt
add world-model synthetic rollouts with reward filtering on top of real rolloutspi0.5 policy · DROID five tasks, multi-tasksuccess rate: +11.6% over baselineImproved
use DSRL noise-space optimization instead of VLAWpi0.5 policy · DROID five tasks, multi-tasksuccess rateNo clear effect
filtered behavior cloning on real successful rollouts onlypi0.5 policy · DROID five tasks, multi-tasksuccess rateImproved
reduce synthetic rollouts from 500 to 250 (ablation)VLAW · Drawing task, 1st roundsuccess rate (%): 68 -> 60Hurt
add real-world success trajectories to policy finetuning with synthetic data (ablation)VLAW · Drawing task, 1st roundsuccess rate (%)Improved
threshold the reward model's yes probabilityVLM reward model · synthetic trajectory filteringfalse-positive trajectoriesImproved

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