AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Towards Practical World Model-based Reinforcement Learning for Vision-Language-Action Models

Zhilong Zhang, Haoxiang Ren, Yihao Sun, Yifei Sheng, Haonan Wang, Zhichao Wu, Haoxin Lin, Pierre-Luc Bacon, Yang Yu

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Vision-based robot RL · 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 interleaved view decoding (ablation)UMM-World · LIBERO-Object wrist viewLPIPS: 0.454 -> 0.254Improved
add interleaved view decoding (ablation)UMM-World · LIBERO-Object head viewLPIPS: 0.116 -> 0.094Improved
add interleaved view decoding (ablation)UMM-World · LIBERO-Object wrist viewPSNR: 13.38 -> 18.76Improved
add interleaved view decoding (ablation)UMM-World · LIBERO-Object rewardF1: 0.799 -> 0.861Improved
add interleaved view decoding (ablation)UMM-World · LIBERO-Object rewardaccuracy: 98.5 -> 98.4No clear effect
add interleaved view decoding (ablation)UMM-World · LIBERO-Object world modelinference time: 8 -> 10Hurt
add pretrained UMM initialization (ablation)UMM-World · LIBERO-Object head viewLPIPS: 0.281 -> 0.094Improved
add pretrained UMM initialization (ablation)UMM-World · LIBERO-Object rewardF1: 0.496 -> 0.861Improved
add pretrained UMM initialization (ablation)UMM-World · LIBERO-Object rewardaccuracy: 94.5 -> 98.4Improved
use 2-chunk branched rollout instead of full-horizon rollout (ablation)VLA-MBPO · LIBERO Longsuccess rate: 52.8 -> 66.8Improved
lengthen branched rollout from 2 to 4 chunks (ablation)VLA-MBPO · LIBERO Longsuccess rate: 66.8 -> 62.9Hurt
shorten branched rollout from 2 to 1 chunk (ablation)VLA-MBPO · LIBERO Longsuccess rate: 66.8 -> 63.9Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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