AI Paper Atlas

Research map / Video / Video segmentation and tracking

Dyn-VPP: Video Prediction Policy Optimization for Improved Visual Dynamics

Zirui Ge, Pengxiang Ding, Baohua Yin, Yemin Wang, Qishen Wang, Zhiyong Xie, Hengtao Li, Runze Suo, Wenxuan Song, Han Zhao, Shangke Lyu, Haoang Li and 5 more

ICML 2026 · Video › Video segmentation and tracking · approach: Stochastic video prediction

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 GRPO post-training of the video prediction model, AGM frozen (ablation)VPP base policy · CALVIN ABC->Daverage trajectory length: 4.28 -> 4.35Improved
add GRPO post-training of VPM, then retrain the action generation model (ablation)VPP base policy · CALVIN ABC->Daverage trajectory length: 4.28 -> 4.56Improved
retrain the AGM after post-training the VPM (versus frozen AGM) (ablation)Dyn-VPP · CALVIN ABC->Daverage trajectory length: 4.35 -> 4.56Improved
add post-training of VPM, AGM frozen (ablation)VPP base policy · CALVIN ABC->D5 tasks in a row (%): 74.7 -> 77.2Improved
add post-training of VPM and AGM (ablation)VPP base policy · CALVIN ABC->D5 tasks in a row (%): 74.7 -> 83.1Improved
add Dyn-VPP post-training (ablation)VPP base policy · CALVIN ABC->Daverage effective rank: 29.28 -> 43.88Improved
add Dyn-VPP post-training (ablation)VPP base policy · CALVIN ABC->Daverage effective rank ratio: 0.0603 -> 0.0814Improved
add Dyn-VPP post-trainingVPP · L-CALVIN ABC->Daverage trajectory length: 5.53 -> 6.73Improved
add Dyn-VPP post-trainingVPP · L-CALVIN ABC->D10 tasks in sequence: 0.28 -> 0.39Improved
use Dyn-VPP instead of Tri-VLA extension of VPPVPP · CALVIN ABC->Daverage trajectory length: 4.37 -> 4.56Improved

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