AI Paper Atlas

Research map / Video / Video generation

Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid Control

Weisheng Xu, Qiwei Wu, Jiaxi Zhang, Jing Tan, Yangfan Li, Yuetong Fang, Jiaqi Xiong, Kai Wu, Rong Ou, Renjing Xu

CVPR 2026 · Video › Video generation · approach: Text-to-motion diffusion · 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
iterate difficulty-escalating data generation with controller feedback (loop0 to loop6)PHC single-primitive tracker · Test sets (2201 clips) averagesuccess rate (%): 55.9 -> 76.9Improved
train on iteratively synthesized data instead of AMASSPHC single-primitive tracker · Test sets (2201 clips) averagesuccess rate (%): 58.3 -> 76.9Improved
apply the iterative framework to MaskedMimic (loop1)MaskedMimic (FC*) · Video-Convertsuccess rate (%): 41.6 -> 62.4Improved
use the first synthesized loop0 data onlyPHC single-primitive tracker · Test sets (2201 clips) averagesuccess rate (%): 58.3 -> 55.9Hurt
add physics tracking metrics to the observation (ablation)CLAIMS feedback loop · EMDB loop3success rate (%): 35.6 -> 64.4Improved
add VLM feedback to the observation (ablation)CLAIMS feedback loop · Four third-party benchmarks, loop3success rate (%): EMDB 60.0 -> 64.4; AIST++ 81.9 -> 82.1; Video-Convert 57.2 -> 50.9Mixed
add the manually curated variable template library (ablation)CLAIMS prompt generation · Four third-party benchmarks, loop3success rate (%): Video-Convert 46.8 -> 50.9; EMDB 53.3 -> 64.4; AIST++ 76.5 -> 82.1Mixed
use feedback-driven iteration instead of one-shot training with equal data size (ablation)PHC single-primitive tracker · AIST++ (1320 clips)success rate (%): 85.3 -> 88.1Improved
use feedback-driven iteration instead of one-shot training with equal data size (ablation)PHC single-primitive tracker · EMDB (45 clips)success rate (%): 73.3 -> 64.4Hurt
raise data difficulty over loopsCLAIMS datasets · PHC+ tracking on own loopssuccess rate (%): 75.3 -> 53.6Improved
use expert variable-library prompts instead of random Gemini prompts for MDM (ablation)MDM synthesis · martial arts t-SNE analysisoverlap with professional motion manifoldImproved

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