AI Paper Atlas

Research map / Video / Video generation

Motion-R1: Enhancing Motion Generation with Decomposed Chain-of-Thought and RL Binding

Runqi Ouyang, Haoyun Li, Zhenyuan Zhang, Xiaofeng Wang, Zeyu Zhang, Zheng Zhu, Guan Huang, Sirui Han, Xingang Wang

ICLR 2026 · Video › Video generation · approach: Text-to-motion diffusion

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 semantic similarity reward to decomposed CoT training (ablation)Motion-R1 · HumanML3DFID: 0.530 -> 0.297Improved
add motion similarity reward to decomposed CoT training (ablation)Motion-R1 · HumanML3DR-Precision Top-1: 0.340 -> 0.483Improved
add the second reward to CoT with one reward (ablation)Motion-R1 · HumanML3DFID: 0.281 -> 0.201Improved
add decomposed CoT to both rewards (ablation)Motion-R1 · HumanML3DR-Precision Top-1: 0.365 -> 0.515Improved
add decomposed CoT to both rewards (ablation)Motion-R1 · HumanML3DFID: 1.656 -> 0.201Improved
add decomposed CoT to both rewards (ablation)Motion-R1 · HumanML3DMModality: 4.008 -> 2.317Hurt
use CoT data engine alone without RL (ablation)Motion-R1 · HumanML3DMModality: 2.317 -> 4.762Improved
end-to-end RL without SFT cold start (ablation)Motion-R1 · training stabilitystabilityHurt
use Motion-R1 instead of MoMaskMotion-R1 · HumanML3DFID: 0.045 -> 0.201Hurt
use Motion-R1 instead of InfiniDreamerMotion-R1 · BABELFID: 1.14 -> 0.53Improved
use Motion-R1 instead of MotionLLMMotion-R1 · KIT-MLFID: 0.781 -> 0.287Improved

Later papers that compared with this method

2 later accepted papers report Motion-R1 as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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